all new onnx model for opencv

.gitignore

@@ -1 +1,3 @@
 ssd_mobilenet_v2_coco_2018_03_29/demo
+
+**/demo

efficientdet-d0/LICENSE

@@ -0,0 +1,203 @@
+Copyright 2020 Google Research.  All rights reserved.
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2020, Google Research.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.	

efficientdet-d0/README.md

@@ -0,0 +1,66 @@
+# EfficientDet-D0
+
+Object detection with EfficientDet-D0 trained on COCO. The model was originally
+distributed as a frozen TensorFlow graph (`efficientdet-d0.pb`) and converted to
+ONNX for use with OpenCV's DNN module. This is a **backbone-only** export: the
+graph emits raw class logits and box regressions, while anchor generation, sigmoid,
+box decoding and non-maximum suppression are performed in host code (see the demos).
+
+## Model Details
+- **Architecture**: EfficientDet-D0
+- **Input**: RGB image, 512×512, raw uint8, NHWC layout (`image_arrays:0`, shape `[1, 512, 512, 3]`)
+- **Output**: raw class logits (`concat:0`, shape `[1, 49104, 90]`) and box regression (`concat_1:0`, shape `[1, 49104, 4]`); anchor decode + NMS are done in host code, not in the graph
+- **Framework**: ONNX (converted from the TensorFlow frozen graph via tf2onnx, opset 18)
+- **Original weights**: https://www.dropbox.com/s/9mqp99fd2tpuqn6/efficientdet-d0.pb?dl=1
+
+The graph outputs are per-anchor predictions only. The demos build the 49104 anchors
+(5 pyramid levels × 9 anchors/cell), apply sigmoid to the logits, decode the box
+regressions relative to the anchors, threshold on confidence and run NMS (IoU 0.6).
+
+## Usage
+
+### Python
+```bash
+python demo.py --model efficientdet-d0_2026jul.onnx --image example_outputs/input_image.png --output example_outputs/output_image.png --conf 0.4
+```
+
+Or import directly:
+```python
+import cv2
+
+net = cv2.dnn.readNet("efficientdet-d0_2026jul.onnx")
+# see demo.py for the full anchor decode + NMS pipeline
+```
+
+### C++
+The C++ demo runs inference with ONNX Runtime (C++ API) and uses OpenCV only for image I/O.
+Install ONNX Runtime (C++) from https://github.com/microsoft/onnxruntime/releases — this build
+uses `onnxruntime-linux-x64-1.25.0` — and adjust the ONNX Runtime and OpenCV paths to your setup:
+```bash
+ORT=/path/to/onnxruntime-linux-x64-1.25.0    # ONNX Runtime release dir (contains include/ and lib/)
+OCV=/path/to/opencv                          # OpenCV source tree
+OCVBUILD=/path/to/opencv/build               # OpenCV build directory (generated headers + libs)
+g++ -std=c++17 demo.cpp -o demo \
+  -I$ORT/include \
+  -I$OCV/include \
+  -I$OCV/modules/core/include \
+  -I$OCV/modules/imgproc/include \
+  -I$OCV/modules/imgcodecs/include \
+  -I$OCVBUILD \
+  -L$ORT/lib -Wl,-rpath,$ORT/lib -lonnxruntime \
+  -L$OCVBUILD/lib -Wl,-rpath,$OCVBUILD/lib -lopencv_imgcodecs -lopencv_imgproc -lopencv_core
+./demo --model efficientdet-d0_2026jul.onnx --image example_outputs/input_image.png --output example_outputs/output_image.png
+```
+
+## Conversion
+The ONNX model was exported from the frozen TensorFlow graph with tf2onnx (opset 18)
+via [convert_to_onnx.py](./convert_to_onnx.py) — input `image_arrays:0`, outputs
+`concat:0` and `concat_1:0`, input shape overridden to `[1, 512, 512, 3]`. Requires
+`tensorflow`, `tf2onnx`, and `onnx`.
+
+```bash
+python convert_to_onnx.py --pb ../pb/efficientdet-d0.pb
+```
+
+## License
+See [LICENSE](./LICENSE) — released under the Apache License 2.0.

efficientdet-d0/convert_to_onnx.py

@@ -0,0 +1,41 @@
+import argparse
+import datetime
+
+import onnx
+import tensorflow as tf
+import tf2onnx
+
+
+def load_graph_def(pb_path):
+    with tf.io.gfile.GFile(pb_path, "rb") as f:
+        graph_def = tf.compat.v1.GraphDef()
+        graph_def.ParseFromString(f.read())
+    return graph_def
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Export efficientdet-d0.pb to ONNX")
+    parser.add_argument("--pb", default="../pb/efficientdet-d0.pb")
+    parser.add_argument("--opset", type=int, default=18)
+    args = parser.parse_args()
+
+    graph_def = load_graph_def(args.pb)
+
+    model_proto, _ = tf2onnx.convert.from_graph_def(
+        graph_def,
+        input_names=["image_arrays:0"],
+        output_names=["concat:0", "concat_1:0"],
+        opset=args.opset,
+        shape_override={"image_arrays:0": [1, 512, 512, 3]},
+    )
+    onnx.checker.check_model(model_proto)
+
+    stamp = datetime.datetime.now().strftime("%Y%b").lower()
+    onnx_path = "efficientdet-d0_%s.onnx" % stamp
+    with open(onnx_path, "wb") as f:
+        f.write(model_proto.SerializeToString())
+    print("wrote", onnx_path)
+
+
+if __name__ == "__main__":
+    main()

efficientdet-d0/demo.cpp

@@ -0,0 +1,144 @@
+#include <onnxruntime_cxx_api.h>
+#include <opencv2/imgproc.hpp>
+#include <opencv2/imgcodecs.hpp>
+#include <algorithm>
+#include <array>
+#include <cmath>
+#include <iostream>
+#include <string>
+#include <vector>
+
+using namespace cv;
+
+static std::string argVal(int argc, char** argv, const std::string& key, const std::string& def)
+{
+    for (int i = 1; i + 1 < argc; ++i)
+        if (key == argv[i]) return argv[i + 1];
+    return def;
+}
+
+struct Det { float x1, y1, x2, y2, score; int cid; };
+
+int main(int argc, char** argv)
+{
+    std::string model  = argVal(argc, argv, "--model",  "efficientdet-d0_2026jul.onnx");
+    std::string image  = argVal(argc, argv, "--image",  "example_outputs/input_image.png");
+    std::string output = argVal(argc, argv, "--output", "example_outputs/output_image.png");
+    float conf = std::stof(argVal(argc, argv, "--conf", "0.4"));
+
+    const int sz = 512;
+
+    Mat img = imread(image);
+    if (img.empty()) { std::cerr << "could not read image: " << image << std::endl; return 1; }
+
+    Mat rgb;
+    cvtColor(img, rgb, COLOR_BGR2RGB);
+    resize(rgb, rgb, Size(sz, sz));
+    if (!rgb.isContinuous()) rgb = rgb.clone();
+
+    Ort::Env env(ORT_LOGGING_LEVEL_WARNING, "demo");
+    Ort::SessionOptions so;
+    Ort::Session session(env, model.c_str(), so);
+    Ort::AllocatorWithDefaultOptions alloc;
+
+    auto in_name = session.GetInputNameAllocated(0, alloc);
+    const char* in_names[] = {in_name.get()};
+
+    size_t nout = session.GetOutputCount();
+    std::vector<Ort::AllocatedStringPtr> out_holders;
+    std::vector<std::string> out_str;
+    for (size_t i = 0; i < nout; ++i)
+    {
+        out_holders.push_back(session.GetOutputNameAllocated(i, alloc));
+        out_str.push_back(out_holders.back().get());
+    }
+    std::vector<const char*> out_names;
+    for (auto& s : out_str) out_names.push_back(s.c_str());
+
+    std::array<int64_t, 4> shape = {1, sz, sz, 3};
+    auto mem = Ort::MemoryInfo::CreateCpu(OrtArenaAllocator, OrtMemTypeDefault);
+    Ort::Value input = Ort::Value::CreateTensor<uint8_t>(mem, rgb.data, (size_t)sz * sz * 3, shape.data(), shape.size());
+
+    auto outs = session.Run(Ort::RunOptions{nullptr}, in_names, &input, 1, out_names.data(), out_names.size());
+
+    const float* boxp = nullptr;
+    const float* clsp = nullptr;
+    int n = 0, nc = 0;
+    for (size_t i = 0; i < outs.size(); ++i)
+    {
+        auto os = outs[i].GetTensorTypeAndShapeInfo().GetShape();
+        const float* p = outs[i].GetTensorMutableData<float>();
+        if (os.back() == 4) { boxp = p; n = (int)os[os.size() - 2]; }
+        else { clsp = p; nc = (int)os.back(); }
+    }
+
+    std::vector<std::array<float, 2>> baseWH;
+    double asp[3][2] = {{1.0, 1.0}, {1.4, 0.7}, {0.7, 1.4}};
+    for (int i = 0; i < 3; ++i) {
+        double s = std::pow(2.0, i / 3.0);
+        for (int a = 0; a < 3; ++a)
+            baseWH.push_back({(float)(32.0 * s * asp[a][0]), (float)(32.0 * s * asp[a][1])});
+    }
+    std::vector<float> acx, acy, aw, ah;
+    for (int lvl = 0; lvl < 5; ++lvl) {
+        int f = sz / (8 << lvl);
+        int step = 8 << lvl;
+        int m = 1 << lvl;
+        for (int y = 0; y < f; ++y)
+            for (int x = 0; x < f; ++x) {
+                float cx = (x + 0.5f) * step;
+                float cy = (y + 0.5f) * step;
+                for (auto& b : baseWH) {
+                    acx.push_back(cx); acy.push_back(cy);
+                    aw.push_back(b[0] * m); ah.push_back(b[1] * m);
+                }
+            }
+    }
+
+    std::vector<Det> dets;
+    for (int a = 0; a < n; ++a) {
+        const float* bp = boxp + (size_t)a * 4;
+        float ycenter = bp[0] * ah[a] + acy[a];
+        float xcenter = bp[1] * aw[a] + acx[a];
+        float bhv = std::exp(bp[2]) * ah[a];
+        float bwv = std::exp(bp[3]) * aw[a];
+        const float* cp = clsp + (size_t)a * nc;
+        int best = 0; float bestLogit = cp[0];
+        for (int c = 1; c < nc; ++c) if (cp[c] > bestLogit) { bestLogit = cp[c]; best = c; }
+        float score = 1.0f / (1.0f + std::exp(-bestLogit));
+        if (score > conf)
+            dets.push_back({(xcenter - bwv / 2) / sz, (ycenter - bhv / 2) / sz,
+                            (xcenter + bwv / 2) / sz, (ycenter + bhv / 2) / sz, score, best});
+    }
+
+    std::sort(dets.begin(), dets.end(), [](const Det& a, const Det& b) { return a.score > b.score; });
+    std::vector<char> removed(dets.size(), 0);
+    std::vector<int> pick;
+    for (size_t i = 0; i < dets.size(); ++i) {
+        if (removed[i]) continue;
+        pick.push_back((int)i);
+        for (size_t j = i + 1; j < dets.size(); ++j) {
+            if (removed[j]) continue;
+            float xx1 = std::max(dets[i].x1, dets[j].x1);
+            float yy1 = std::max(dets[i].y1, dets[j].y1);
+            float xx2 = std::min(dets[i].x2, dets[j].x2);
+            float yy2 = std::min(dets[i].y2, dets[j].y2);
+            float inter = std::max(0.0f, xx2 - xx1) * std::max(0.0f, yy2 - yy1);
+            float ai = (dets[i].x2 - dets[i].x1) * (dets[i].y2 - dets[i].y1);
+            float aj = (dets[j].x2 - dets[j].x1) * (dets[j].y2 - dets[j].y1);
+            if (inter / (ai + aj - inter + 1e-9f) > 0.6f) removed[j] = 1;
+        }
+    }
+
+    std::cout << "efficientdet-d0 " << pick.size() << " detections" << std::endl;
+    int w = img.cols, h = img.rows;
+    for (int idx : pick) {
+        const Det& d = dets[idx];
+        std::cout << format("%d %.3f %.3f %.3f %.3f %.3f", d.cid, d.score, d.x1, d.y1, d.x2, d.y2) << std::endl;
+        rectangle(img, Point((int)(d.x1 * w), (int)(d.y1 * h)), Point((int)(d.x2 * w), (int)(d.y2 * h)), Scalar(0, 255, 0), 2);
+        putText(img, format("%d:%.2f", d.cid, d.score), Point((int)(d.x1 * w), (int)(d.y1 * h) - 5), FONT_HERSHEY_SIMPLEX, 0.5, Scalar(0, 255, 0), 1);
+    }
+    imwrite(output, img);
+    std::cout << "wrote " << output << std::endl;
+    return 0;
+}

efficientdet-d0/demo.py

@@ -0,0 +1,105 @@
+import argparse
+import glob
+import os
+
+import cv2 as cv
+import numpy as np
+import onnxruntime as ort
+
+here = os.path.dirname(os.path.abspath(__file__))
+sz = 512
+
+
+def build_anchors():
+    scales = [2.0 ** (i / 3.0) for i in range(3)]
+    aspects = [(1.0, 1.0), (1.4, 0.7), (0.7, 1.4)]
+    base = []
+    for s in scales:
+        for aw, ah in aspects:
+            base.append((32.0 * s * aw, 32.0 * s * ah))
+    anchors = []
+    for lvl in range(5):
+        f = sz // (8 * 2 ** lvl)
+        step = 8 * 2 ** lvl
+        m = 2 ** lvl
+        for y in range(f):
+            for x in range(f):
+                cx = (x + 0.5) * step
+                cy = (y + 0.5) * step
+                for bw, bh in base:
+                    anchors.append((cx, cy, bw * m, bh * m))
+    return np.array(anchors, np.float32)
+
+
+def main():
+    parser = argparse.ArgumentParser(description="EfficientDet-D0 (ONNX) object detection demo")
+    parser.add_argument("--model", default=None)
+    parser.add_argument("--image", default=os.path.join(here, "example_outputs", "input_image.png"))
+    parser.add_argument("--output", default=os.path.join(here, "example_outputs", "output_image.png"))
+    parser.add_argument("--conf", type=float, default=0.4)
+    args = parser.parse_args()
+
+    model = args.model
+    if model is None:
+        found = glob.glob(os.path.join(here, "*.onnx"))
+        if not found:
+            raise SystemExit("no onnx, run convert_to_onnx.py")
+        model = found[0]
+
+    img = cv.imread(args.image)
+    if img is None:
+        raise SystemExit("could not read image: %s" % args.image)
+
+    anchors = build_anchors()
+    acx, acy, aw, ah = anchors[:, 0], anchors[:, 1], anchors[:, 2], anchors[:, 3]
+
+    sess = ort.InferenceSession(model, providers=["CPUExecutionProvider"])
+    inp = cv.resize(cv.cvtColor(img, cv.COLOR_BGR2RGB), (sz, sz))
+    res = sess.run(None, {sess.get_inputs()[0].name: inp[None].astype(np.uint8)})
+    box = next(a for a in res if a.shape[-1] == 4).reshape(-1, 4)
+    cls = next(a for a in res if a.shape[-1] != 4).reshape(box.shape[0], -1)
+
+    ycenter = box[:, 0] * ah + acy
+    xcenter = box[:, 1] * aw + acx
+    bh = np.exp(box[:, 2]) * ah
+    bw = np.exp(box[:, 3]) * aw
+    boxes = np.stack([xcenter - bw / 2, ycenter - bh / 2, xcenter + bw / 2, ycenter + bh / 2], 1) / sz
+
+    prob = 1.0 / (1.0 + np.exp(-cls))
+    cid = prob.argmax(1)
+    scores = prob.max(1)
+
+    keep = scores > args.conf
+    boxes = boxes[keep]
+    scores = scores[keep]
+    cid = cid[keep]
+    order = scores.argsort()[::-1]
+    pick = []
+    while order.size:
+        i = order[0]
+        pick.append(i)
+        xx1 = np.maximum(boxes[i, 0], boxes[order[1:], 0])
+        yy1 = np.maximum(boxes[i, 1], boxes[order[1:], 1])
+        xx2 = np.minimum(boxes[i, 2], boxes[order[1:], 2])
+        yy2 = np.minimum(boxes[i, 3], boxes[order[1:], 3])
+        iw = np.maximum(0, xx2 - xx1)
+        ih = np.maximum(0, yy2 - yy1)
+        inter = iw * ih
+        ai = (boxes[i, 2] - boxes[i, 0]) * (boxes[i, 3] - boxes[i, 1])
+        aj = (boxes[order[1:], 2] - boxes[order[1:], 0]) * (boxes[order[1:], 3] - boxes[order[1:], 1])
+        iou = inter / (ai + aj - inter + 1e-9)
+        order = order[1:][iou <= 0.6]
+
+    print("efficientdet-d0", len(pick), "detections")
+    h, w = img.shape[:2]
+    for i in pick:
+        x1, y1, x2, y2 = boxes[i]
+        print(int(cid[i]), round(float(scores[i]), 3), round(float(x1), 3), round(float(y1), 3), round(float(x2), 3), round(float(y2), 3))
+        cv.rectangle(img, (int(x1 * w), int(y1 * h)), (int(x2 * w), int(y2 * h)), (0, 255, 0), 2)
+        cv.putText(img, "%d:%.2f" % (int(cid[i]), scores[i]), (int(x1 * w), int(y1 * h) - 5), cv.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 0), 1)
+    cv.imwrite(args.output, img)
+    print("wrote", args.output)
+
+
+if __name__ == "__main__":
+    main()

efficientdet-d0/efficientdet-d0_2026jul.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db344f69adf1c529e08a36bbaa1779d98b4b81621a2f046656fb966bdfdc6298
+size 15671001

faster_rcnn_inception_v2_coco_2018_01_28/LICENSE

@@ -0,0 +1,203 @@
+Copyright 2015 The TensorFlow Authors.  All rights reserved.
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2015, The TensorFlow Authors.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+limitations under the License.

faster_rcnn_inception_v2_coco_2018_01_28/README.md

@@ -0,0 +1,53 @@
+# Faster-RCNN InceptionV2 (COCO)
+
+Object detection with the Faster-RCNN meta-architecture and an Inception v2 backbone,
+trained on the COCO dataset. The model was originally distributed as a frozen TensorFlow
+graph (`faster_rcnn_inception_v2_coco_2018_01_28.pb`) from the TensorFlow Object Detection
+API and converted to ONNX for inference with ONNX Runtime.
+
+## Model Details
+- **Architecture**: Faster-RCNN with an Inception v2 backbone
+- **Input**: RGB image, uint8, NHWC layout (`image_tensor:0`, shape `[1, H, W, 3]`)
+- **Output**: `detection_boxes:0` (normalized `[ymin, xmin, ymax, xmax]`), `detection_scores:0`, `detection_classes:0` (1-based COCO ids), `num_detections:0`
+- **Framework**: ONNX (converted from the TensorFlow frozen graph via tf2onnx, opset 18)
+- **Original weights**: http://download.tensorflow.org/models/object_detection/faster_rcnn_inception_v2_coco_2018_01_28.tar.gz
+
+## Usage
+
+### Python
+```bash
+python demo.py --model faster_rcnn_inception_v2_coco_2018_01_28_2026jul.onnx --image example_outputs/input_image.png --output example_outputs/output_image.png --conf 0.3
+```
+
+### C++
+The C++ demo runs inference with ONNX Runtime (C++ API) and uses OpenCV only for image I/O.
+Install ONNX Runtime (C++) from https://github.com/microsoft/onnxruntime/releases — this build
+uses `onnxruntime-linux-x64-1.25.0` — and adjust the ONNX Runtime and OpenCV paths to your setup:
+```bash
+ORT=/path/to/onnxruntime-linux-x64-1.25.0    # ONNX Runtime release dir (contains include/ and lib/)
+OCV=/path/to/opencv                          # OpenCV source tree
+OCVBUILD=/path/to/opencv/build               # OpenCV build directory (generated headers + libs)
+g++ -std=c++17 demo.cpp -o demo \
+  -I$ORT/include \
+  -I$OCV/include \
+  -I$OCV/modules/core/include \
+  -I$OCV/modules/imgproc/include \
+  -I$OCV/modules/imgcodecs/include \
+  -I$OCVBUILD \
+  -L$ORT/lib -Wl,-rpath,$ORT/lib -lonnxruntime \
+  -L$OCVBUILD/lib -Wl,-rpath,$OCVBUILD/lib -lopencv_imgcodecs -lopencv_imgproc -lopencv_core
+./demo --model faster_rcnn_inception_v2_coco_2018_01_28_2026jul.onnx --image example_outputs/input_image.png --output example_outputs/output_image.png
+```
+
+## Conversion
+The ONNX model was exported from the frozen TensorFlow graph with tf2onnx (opset 18)
+via [convert_to_onnx.py](./convert_to_onnx.py) — input `image_tensor:0`, outputs
+`detection_boxes:0`, `detection_scores:0`, `detection_classes:0`, `num_detections:0`.
+Requires `tensorflow`, `tf2onnx`, and `onnx`.
+
+```bash
+python convert_to_onnx.py --pb ../pb/faster_rcnn_inception_v2_coco_2018_01_28.pb
+```
+
+## License
+See [LICENSE](./LICENSE) — the model is released by the TensorFlow Authors under the Apache License 2.0.

faster_rcnn_inception_v2_coco_2018_01_28/convert_to_onnx.py

@@ -0,0 +1,40 @@
+import argparse
+import datetime
+
+import onnx
+import tensorflow as tf
+import tf2onnx
+
+
+def load_graph_def(pb_path):
+    with tf.io.gfile.GFile(pb_path, "rb") as f:
+        graph_def = tf.compat.v1.GraphDef()
+        graph_def.ParseFromString(f.read())
+    return graph_def
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Export faster_rcnn_inception_v2_coco_2018_01_28.pb to ONNX")
+    parser.add_argument("--pb", default="../pb/faster_rcnn_inception_v2_coco_2018_01_28.pb")
+    parser.add_argument("--opset", type=int, default=18)
+    args = parser.parse_args()
+
+    graph_def = load_graph_def(args.pb)
+
+    model_proto, _ = tf2onnx.convert.from_graph_def(
+        graph_def,
+        input_names=["image_tensor:0"],
+        output_names=["detection_boxes:0", "detection_scores:0", "detection_classes:0", "num_detections:0"],
+        opset=args.opset,
+    )
+    onnx.checker.check_model(model_proto)
+
+    stamp = datetime.datetime.now().strftime("%Y%b").lower()
+    onnx_path = "faster_rcnn_inception_v2_coco_2018_01_28_%s.onnx" % stamp
+    with open(onnx_path, "wb") as f:
+        f.write(model_proto.SerializeToString())
+    print("wrote", onnx_path)
+
+
+if __name__ == "__main__":
+    main()

faster_rcnn_inception_v2_coco_2018_01_28/demo.cpp

@@ -0,0 +1,100 @@
+#include <onnxruntime_cxx_api.h>
+#include <opencv2/imgproc.hpp>
+#include <opencv2/imgcodecs.hpp>
+#include <array>
+#include <cstdint>
+#include <iostream>
+#include <string>
+#include <vector>
+
+static std::string argVal(int argc, char** argv, const std::string& key, const std::string& def)
+{
+    for (int i = 1; i + 1 < argc; ++i)
+        if (key == argv[i]) return argv[i + 1];
+    return def;
+}
+
+int main(int argc, char** argv)
+{
+    std::string model  = argVal(argc, argv, "--model",  "faster_rcnn_inception_v2_coco_2018_01_28_2026jul.onnx");
+    std::string image  = argVal(argc, argv, "--image",  "example_outputs/input_image.png");
+    std::string output = argVal(argc, argv, "--output", "example_outputs/output_image.png");
+    float conf = std::stof(argVal(argc, argv, "--conf", "0.3"));
+
+    cv::Mat img = cv::imread(image);
+    if (img.empty())
+    {
+        std::cerr << "could not read image: " << image << std::endl;
+        return 1;
+    }
+
+    const int W = 800, H = 600;
+    cv::Mat rgb;
+    cv::cvtColor(img, rgb, cv::COLOR_BGR2RGB);
+    cv::resize(rgb, rgb, cv::Size(W, H));
+    if (!rgb.isContinuous()) rgb = rgb.clone();
+
+    Ort::Env env(ORT_LOGGING_LEVEL_WARNING, "demo");
+    Ort::SessionOptions so;
+    Ort::Session session(env, model.c_str(), so);
+    Ort::AllocatorWithDefaultOptions alloc;
+
+    auto in_name = session.GetInputNameAllocated(0, alloc);
+    const char* in_names[] = {in_name.get()};
+
+    size_t nout = session.GetOutputCount();
+    std::vector<Ort::AllocatedStringPtr> out_holders;
+    std::vector<std::string> out_strs;
+    for (size_t i = 0; i < nout; ++i)
+    {
+        out_holders.push_back(session.GetOutputNameAllocated(i, alloc));
+        out_strs.push_back(out_holders.back().get());
+    }
+    std::vector<const char*> out_names;
+    for (auto& s : out_strs) out_names.push_back(s.c_str());
+
+    std::array<int64_t, 4> shape = {1, H, W, 3};
+    auto mem = Ort::MemoryInfo::CreateCpu(OrtArenaAllocator, OrtMemTypeDefault);
+    Ort::Value input = Ort::Value::CreateTensor<uint8_t>(mem, rgb.data, (size_t)H * W * 3, shape.data(), shape.size());
+
+    auto outs = session.Run(Ort::RunOptions{nullptr}, in_names, &input, 1, out_names.data(), out_names.size());
+
+    float* boxes = nullptr;
+    float* scores = nullptr;
+    float* classes = nullptr;
+    float* numd = nullptr;
+    for (size_t i = 0; i < nout; ++i)
+    {
+        float* p = outs[i].GetTensorMutableData<float>();
+        const std::string& n = out_strs[i];
+        if (n.find("detection_boxes") != std::string::npos) boxes = p;
+        else if (n.find("detection_scores") != std::string::npos) scores = p;
+        else if (n.find("detection_classes") != std::string::npos) classes = p;
+        else if (n.find("num_detections") != std::string::npos) numd = p;
+    }
+    int nd = (int)numd[0];
+
+    int w = img.cols, h = img.rows;
+    std::vector<int> kept;
+    for (int i = 0; i < nd; ++i)
+        if (scores[i] >= conf) kept.push_back(i);
+
+    std::cout << "faster_rcnn_inception_v2_coco_2018_01_28 " << kept.size() << " detections" << std::endl;
+    for (int i : kept)
+    {
+        int cls = (int)classes[i] - 1;
+        float score = scores[i];
+        float ymin = boxes[i * 4 + 0], xmin = boxes[i * 4 + 1];
+        float ymax = boxes[i * 4 + 2], xmax = boxes[i * 4 + 3];
+        cv::Point p1((int)(xmin * w), (int)(ymin * h));
+        cv::Point p2((int)(xmax * w), (int)(ymax * h));
+        cv::rectangle(img, p1, p2, cv::Scalar(0, 255, 0), 2);
+        cv::putText(img, cv::format("%d:%.2f", cls, score), cv::Point(p1.x, p1.y - 5),
+                    cv::FONT_HERSHEY_SIMPLEX, 0.5, cv::Scalar(0, 255, 0), 1);
+        std::cout << cls << " " << cv::format("%.3f %.3f %.3f %.3f %.3f", score, xmin, ymin, xmax, ymax) << std::endl;
+    }
+
+    cv::imwrite(output, img);
+    std::cout << "wrote " << output << std::endl;
+    return 0;
+}

faster_rcnn_inception_v2_coco_2018_01_28/demo.py

@@ -0,0 +1,51 @@
+import argparse
+import glob
+import os
+
+import cv2 as cv
+import numpy as np
+import onnxruntime as ort
+
+here = os.path.dirname(os.path.abspath(__file__))
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Faster-RCNN InceptionV2 (COCO) ONNX detection demo")
+    parser.add_argument("--model", default=None)
+    parser.add_argument("--image", default=os.path.join(here, "example_outputs", "input_image.png"))
+    parser.add_argument("--output", default=os.path.join(here, "example_outputs", "output_image.png"))
+    parser.add_argument("--conf", type=float, default=0.3)
+    args = parser.parse_args()
+
+    model = args.model or glob.glob(os.path.join(here, "*.onnx"))[0]
+    img = cv.imread(args.image)
+    if img is None:
+        raise SystemExit("could not read image: %s" % args.image)
+
+    rgb = cv.resize(cv.cvtColor(img, cv.COLOR_BGR2RGB), (800, 600))
+    sess = ort.InferenceSession(model, providers=["CPUExecutionProvider"])
+    res = sess.run(None, {sess.get_inputs()[0].name: rgb[None].astype(np.uint8)})
+    onames = [o.name for o in sess.get_outputs()]
+    boxes = res[[i for i, n in enumerate(onames) if "detection_boxes" in n][0]].reshape(-1, 4)
+    scores = res[[i for i, n in enumerate(onames) if "detection_scores" in n][0]].reshape(-1)
+    classes = res[[i for i, n in enumerate(onames) if "detection_classes" in n][0]].reshape(-1)
+    nd = int(res[[i for i, n in enumerate(onames) if "num_detections" in n][0]].reshape(-1)[0])
+
+    out = np.zeros((nd, 7), np.float32)
+    out[:, 1] = classes[:nd] - 1
+    out[:, 2] = scores[:nd]
+    out[:, 3:7] = boxes[:nd][:, [1, 0, 3, 2]]
+
+    h, w = img.shape[:2]
+    kept = [row for row in out if row[2] >= args.conf]
+    print(os.path.basename(here), len(kept), "detections")
+    for row in kept:
+        cv.rectangle(img, (int(row[3] * w), int(row[4] * h)), (int(row[5] * w), int(row[6] * h)), (0, 255, 0), 2)
+        cv.putText(img, "%d:%.2f" % (int(row[1]), row[2]), (int(row[3] * w), int(row[4] * h) - 5), cv.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 0), 1)
+        print(int(row[1]), round(float(row[2]), 3), round(float(row[3]), 3), round(float(row[4]), 3), round(float(row[5]), 3), round(float(row[6]), 3))
+    cv.imwrite(args.output, img)
+    print("wrote", args.output)
+
+
+if __name__ == "__main__":
+    main()

faster_rcnn_inception_v2_coco_2018_01_28/faster_rcnn_inception_v2_coco_2018_01_28_2026jul.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bcf541da5a58e9d8ab6c16e4c803ccc4e0da7dd62ad89311ae2a75c36b39835b
+size 57016094

faster_rcnn_resnet50_coco_2018_01_28/LICENSE

@@ -0,0 +1,203 @@
+Copyright 2015 The TensorFlow Authors.  All rights reserved.
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2015, The TensorFlow Authors.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+limitations under the License.

faster_rcnn_resnet50_coco_2018_01_28/README.md

@@ -0,0 +1,53 @@
+# Faster-RCNN ResNet-50 (COCO)
+
+Object detection with the Faster-RCNN meta-architecture and a ResNet-50 backbone,
+trained on the COCO dataset. The model was originally distributed as a frozen TensorFlow
+graph (`faster_rcnn_resnet50_coco_2018_01_28.pb`) from the TensorFlow Object Detection
+API and converted to ONNX for inference with ONNX Runtime.
+
+## Model Details
+- **Architecture**: Faster-RCNN with a ResNet-50 backbone
+- **Input**: RGB image, uint8, NHWC layout (`image_tensor:0`, shape `[1, H, W, 3]`)
+- **Output**: `detection_boxes:0` (normalized `[ymin, xmin, ymax, xmax]`), `detection_scores:0`, `detection_classes:0` (1-based COCO ids), `num_detections:0`
+- **Framework**: ONNX (converted from the TensorFlow frozen graph via tf2onnx, opset 18)
+- **Original weights**: http://download.tensorflow.org/models/object_detection/faster_rcnn_resnet50_coco_2018_01_28.tar.gz
+
+## Usage
+
+### Python
+```bash
+python demo.py --model faster_rcnn_resnet50_coco_2018_01_28_2026jul.onnx --image example_outputs/input_image.png --output example_outputs/output_image.png --conf 0.3
+```
+
+### C++
+The C++ demo runs inference with ONNX Runtime (C++ API) and uses OpenCV only for image I/O.
+Install ONNX Runtime (C++) from https://github.com/microsoft/onnxruntime/releases — this build
+uses `onnxruntime-linux-x64-1.25.0` — and adjust the ONNX Runtime and OpenCV paths to your setup:
+```bash
+ORT=/path/to/onnxruntime-linux-x64-1.25.0    # ONNX Runtime release dir (contains include/ and lib/)
+OCV=/path/to/opencv                          # OpenCV source tree
+OCVBUILD=/path/to/opencv/build               # OpenCV build directory (generated headers + libs)
+g++ -std=c++17 demo.cpp -o demo \
+  -I$ORT/include \
+  -I$OCV/include \
+  -I$OCV/modules/core/include \
+  -I$OCV/modules/imgproc/include \
+  -I$OCV/modules/imgcodecs/include \
+  -I$OCVBUILD \
+  -L$ORT/lib -Wl,-rpath,$ORT/lib -lonnxruntime \
+  -L$OCVBUILD/lib -Wl,-rpath,$OCVBUILD/lib -lopencv_imgcodecs -lopencv_imgproc -lopencv_core
+./demo --model faster_rcnn_resnet50_coco_2018_01_28_2026jul.onnx --image example_outputs/input_image.png --output example_outputs/output_image.png
+```
+
+## Conversion
+The ONNX model was exported from the frozen TensorFlow graph with tf2onnx (opset 18)
+via [convert_to_onnx.py](./convert_to_onnx.py) — input `image_tensor:0`, outputs
+`detection_boxes:0`, `detection_scores:0`, `detection_classes:0`, `num_detections:0`.
+Requires `tensorflow`, `tf2onnx`, and `onnx`.
+
+```bash
+python convert_to_onnx.py --pb ../pb/faster_rcnn_resnet50_coco_2018_01_28.pb
+```
+
+## License
+See [LICENSE](./LICENSE) — the model is released by the TensorFlow Authors under the Apache License 2.0.

faster_rcnn_resnet50_coco_2018_01_28/convert_to_onnx.py

@@ -0,0 +1,40 @@
+import argparse
+import datetime
+
+import onnx
+import tensorflow as tf
+import tf2onnx
+
+
+def load_graph_def(pb_path):
+    with tf.io.gfile.GFile(pb_path, "rb") as f:
+        graph_def = tf.compat.v1.GraphDef()
+        graph_def.ParseFromString(f.read())
+    return graph_def
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Export faster_rcnn_resnet50_coco_2018_01_28.pb to ONNX")
+    parser.add_argument("--pb", default="../pb/faster_rcnn_resnet50_coco_2018_01_28.pb")
+    parser.add_argument("--opset", type=int, default=18)
+    args = parser.parse_args()
+
+    graph_def = load_graph_def(args.pb)
+
+    model_proto, _ = tf2onnx.convert.from_graph_def(
+        graph_def,
+        input_names=["image_tensor:0"],
+        output_names=["detection_boxes:0", "detection_scores:0", "detection_classes:0", "num_detections:0"],
+        opset=args.opset,
+    )
+    onnx.checker.check_model(model_proto)
+
+    stamp = datetime.datetime.now().strftime("%Y%b").lower()
+    onnx_path = "faster_rcnn_resnet50_coco_2018_01_28_%s.onnx" % stamp
+    with open(onnx_path, "wb") as f:
+        f.write(model_proto.SerializeToString())
+    print("wrote", onnx_path)
+
+
+if __name__ == "__main__":
+    main()

faster_rcnn_resnet50_coco_2018_01_28/demo.cpp

@@ -0,0 +1,100 @@
+#include <onnxruntime_cxx_api.h>
+#include <opencv2/imgproc.hpp>
+#include <opencv2/imgcodecs.hpp>
+#include <array>
+#include <cstdint>
+#include <iostream>
+#include <string>
+#include <vector>
+
+static std::string argVal(int argc, char** argv, const std::string& key, const std::string& def)
+{
+    for (int i = 1; i + 1 < argc; ++i)
+        if (key == argv[i]) return argv[i + 1];
+    return def;
+}
+
+int main(int argc, char** argv)
+{
+    std::string model  = argVal(argc, argv, "--model",  "faster_rcnn_resnet50_coco_2018_01_28_2026jul.onnx");
+    std::string image  = argVal(argc, argv, "--image",  "example_outputs/input_image.png");
+    std::string output = argVal(argc, argv, "--output", "example_outputs/output_image.png");
+    float conf = std::stof(argVal(argc, argv, "--conf", "0.3"));
+
+    cv::Mat img = cv::imread(image);
+    if (img.empty())
+    {
+        std::cerr << "could not read image: " << image << std::endl;
+        return 1;
+    }
+
+    const int W = 800, H = 600;
+    cv::Mat rgb;
+    cv::cvtColor(img, rgb, cv::COLOR_BGR2RGB);
+    cv::resize(rgb, rgb, cv::Size(W, H));
+    if (!rgb.isContinuous()) rgb = rgb.clone();
+
+    Ort::Env env(ORT_LOGGING_LEVEL_WARNING, "demo");
+    Ort::SessionOptions so;
+    Ort::Session session(env, model.c_str(), so);
+    Ort::AllocatorWithDefaultOptions alloc;
+
+    auto in_name = session.GetInputNameAllocated(0, alloc);
+    const char* in_names[] = {in_name.get()};
+
+    size_t nout = session.GetOutputCount();
+    std::vector<Ort::AllocatedStringPtr> out_holders;
+    std::vector<std::string> out_strs;
+    for (size_t i = 0; i < nout; ++i)
+    {
+        out_holders.push_back(session.GetOutputNameAllocated(i, alloc));
+        out_strs.push_back(out_holders.back().get());
+    }
+    std::vector<const char*> out_names;
+    for (auto& s : out_strs) out_names.push_back(s.c_str());
+
+    std::array<int64_t, 4> shape = {1, H, W, 3};
+    auto mem = Ort::MemoryInfo::CreateCpu(OrtArenaAllocator, OrtMemTypeDefault);
+    Ort::Value input = Ort::Value::CreateTensor<uint8_t>(mem, rgb.data, (size_t)H * W * 3, shape.data(), shape.size());
+
+    auto outs = session.Run(Ort::RunOptions{nullptr}, in_names, &input, 1, out_names.data(), out_names.size());
+
+    float* boxes = nullptr;
+    float* scores = nullptr;
+    float* classes = nullptr;
+    float* numd = nullptr;
+    for (size_t i = 0; i < nout; ++i)
+    {
+        float* p = outs[i].GetTensorMutableData<float>();
+        const std::string& n = out_strs[i];
+        if (n.find("detection_boxes") != std::string::npos) boxes = p;
+        else if (n.find("detection_scores") != std::string::npos) scores = p;
+        else if (n.find("detection_classes") != std::string::npos) classes = p;
+        else if (n.find("num_detections") != std::string::npos) numd = p;
+    }
+    int nd = (int)numd[0];
+
+    int w = img.cols, h = img.rows;
+    std::vector<int> kept;
+    for (int i = 0; i < nd; ++i)
+        if (scores[i] >= conf) kept.push_back(i);
+
+    std::cout << "faster_rcnn_resnet50_coco_2018_01_28 " << kept.size() << " detections" << std::endl;
+    for (int i : kept)
+    {
+        int cls = (int)classes[i] - 1;
+        float score = scores[i];
+        float ymin = boxes[i * 4 + 0], xmin = boxes[i * 4 + 1];
+        float ymax = boxes[i * 4 + 2], xmax = boxes[i * 4 + 3];
+        cv::Point p1((int)(xmin * w), (int)(ymin * h));
+        cv::Point p2((int)(xmax * w), (int)(ymax * h));
+        cv::rectangle(img, p1, p2, cv::Scalar(0, 255, 0), 2);
+        cv::putText(img, cv::format("%d:%.2f", cls, score), cv::Point(p1.x, p1.y - 5),
+                    cv::FONT_HERSHEY_SIMPLEX, 0.5, cv::Scalar(0, 255, 0), 1);
+        std::cout << cls << " " << cv::format("%.3f %.3f %.3f %.3f %.3f", score, xmin, ymin, xmax, ymax) << std::endl;
+    }
+
+    cv::imwrite(output, img);
+    std::cout << "wrote " << output << std::endl;
+    return 0;
+}

faster_rcnn_resnet50_coco_2018_01_28/demo.py

@@ -0,0 +1,51 @@
+import argparse
+import glob
+import os
+
+import cv2 as cv
+import numpy as np
+import onnxruntime as ort
+
+here = os.path.dirname(os.path.abspath(__file__))
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Faster-RCNN ResNet-50 (COCO) ONNX detection demo")
+    parser.add_argument("--model", default=None)
+    parser.add_argument("--image", default=os.path.join(here, "example_outputs", "input_image.png"))
+    parser.add_argument("--output", default=os.path.join(here, "example_outputs", "output_image.png"))
+    parser.add_argument("--conf", type=float, default=0.3)
+    args = parser.parse_args()
+
+    model = args.model or glob.glob(os.path.join(here, "*.onnx"))[0]
+    img = cv.imread(args.image)
+    if img is None:
+        raise SystemExit("could not read image: %s" % args.image)
+
+    rgb = cv.resize(cv.cvtColor(img, cv.COLOR_BGR2RGB), (800, 600))
+    sess = ort.InferenceSession(model, providers=["CPUExecutionProvider"])
+    res = sess.run(None, {sess.get_inputs()[0].name: rgb[None].astype(np.uint8)})
+    onames = [o.name for o in sess.get_outputs()]
+    boxes = res[[i for i, n in enumerate(onames) if "detection_boxes" in n][0]].reshape(-1, 4)
+    scores = res[[i for i, n in enumerate(onames) if "detection_scores" in n][0]].reshape(-1)
+    classes = res[[i for i, n in enumerate(onames) if "detection_classes" in n][0]].reshape(-1)
+    nd = int(res[[i for i, n in enumerate(onames) if "num_detections" in n][0]].reshape(-1)[0])
+
+    out = np.zeros((nd, 7), np.float32)
+    out[:, 1] = classes[:nd] - 1
+    out[:, 2] = scores[:nd]
+    out[:, 3:7] = boxes[:nd][:, [1, 0, 3, 2]]
+
+    h, w = img.shape[:2]
+    kept = [row for row in out if row[2] >= args.conf]
+    print(os.path.basename(here), len(kept), "detections")
+    for row in kept:
+        cv.rectangle(img, (int(row[3] * w), int(row[4] * h)), (int(row[5] * w), int(row[6] * h)), (0, 255, 0), 2)
+        cv.putText(img, "%d:%.2f" % (int(row[1]), row[2]), (int(row[3] * w), int(row[4] * h) - 5), cv.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 0), 1)
+        print(int(row[1]), round(float(row[2]), 3), round(float(row[3]), 3), round(float(row[4]), 3), round(float(row[5]), 3), round(float(row[6]), 3))
+    cv.imwrite(args.output, img)
+    print("wrote", args.output)
+
+
+if __name__ == "__main__":
+    main()

faster_rcnn_resnet50_coco_2018_01_28/faster_rcnn_resnet50_coco_2018_01_28_2026jul.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:217613103b36eba4771087bdae63a7c04ed545e7850087f205990db7e5f18b88
+size 120414387

mask_rcnn_inception_v2_coco_2018_01_28/LICENSE

@@ -0,0 +1,203 @@
+Copyright 2015 The TensorFlow Authors.  All rights reserved.
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2015, The TensorFlow Authors.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+limitations under the License.

mask_rcnn_inception_v2_coco_2018_01_28/README.md

@@ -0,0 +1,55 @@
+# Mask-RCNN Inception v2 COCO
+
+Instance segmentation with the Mask-RCNN Inception v2 network trained on the COCO dataset.
+The model was originally distributed as a frozen TensorFlow graph
+(`mask_rcnn_inception_v2_coco_2018_01_28.pb`) from the TensorFlow Object Detection API
+and converted to ONNX for use with ONNX Runtime.
+
+## Model Details
+- **Architecture**: Mask-RCNN with an Inception v2 backbone
+- **Input**: RGB image, uint8, NHWC layout (`image_tensor:0`); the demo resizes to 800×800
+- **Output**: `num_detections:0`, `detection_boxes:0` (normalized `[ymin, xmin, ymax, xmax]`),
+  `detection_scores:0`, `detection_classes:0` (COCO ids, subtract 1 for a 0-based label),
+  and `detection_masks:0` (a 15×15 mask per detection, resized to its box)
+- **Framework**: ONNX (converted from the TensorFlow frozen graph via tf2onnx, opset 18)
+- **Original weights**: http://download.tensorflow.org/models/object_detection/mask_rcnn_inception_v2_coco_2018_01_28.tar.gz
+
+## Usage
+
+### Python
+```bash
+python demo.py --model mask_rcnn_inception_v2_coco_2018_01_28_2026jul.onnx --image example_outputs/input_image.png --output example_outputs/output_image.png --conf 0.3
+```
+
+### C++
+The C++ demo runs inference with ONNX Runtime (C++ API) and uses OpenCV only for image I/O.
+Install ONNX Runtime (C++) from https://github.com/microsoft/onnxruntime/releases — this build
+uses `onnxruntime-linux-x64-1.25.0` — and adjust the ONNX Runtime and OpenCV paths to your setup:
+```bash
+ORT=/path/to/onnxruntime-linux-x64-1.25.0    # ONNX Runtime release dir (contains include/ and lib/)
+OCV=/path/to/opencv                          # OpenCV source tree
+OCVBUILD=/path/to/opencv/build               # OpenCV build directory (generated headers + libs)
+g++ -std=c++17 demo.cpp -o demo \
+  -I$ORT/include \
+  -I$OCV/include \
+  -I$OCV/modules/core/include \
+  -I$OCV/modules/imgproc/include \
+  -I$OCV/modules/imgcodecs/include \
+  -I$OCVBUILD \
+  -L$ORT/lib -Wl,-rpath,$ORT/lib -lonnxruntime \
+  -L$OCVBUILD/lib -Wl,-rpath,$OCVBUILD/lib -lopencv_imgcodecs -lopencv_imgproc -lopencv_core
+./demo --model mask_rcnn_inception_v2_coco_2018_01_28_2026jul.onnx --image example_outputs/input_image.png --output example_outputs/output_image.png
+```
+
+## Conversion
+The ONNX model was exported from the frozen TensorFlow graph with tf2onnx (opset 18)
+via [convert_to_onnx.py](./convert_to_onnx.py) — input `image_tensor:0`, outputs
+`num_detections:0`, `detection_boxes:0`, `detection_scores:0`, `detection_classes:0`,
+and `detection_masks:0`. Requires `tensorflow`, `tf2onnx`, and `onnx`.
+
+```bash
+python convert_to_onnx.py --pb ../pb/mask_rcnn_inception_v2_coco_2018_01_28.pb
+```
+
+## License
+See [LICENSE](./LICENSE) — the model is released by the TensorFlow Authors under the Apache License 2.0.

mask_rcnn_inception_v2_coco_2018_01_28/convert_to_onnx.py

@@ -0,0 +1,46 @@
+import argparse
+import datetime
+
+import onnx
+import tensorflow as tf
+import tf2onnx
+
+
+def load_graph_def(pb_path):
+    with tf.io.gfile.GFile(pb_path, "rb") as f:
+        graph_def = tf.compat.v1.GraphDef()
+        graph_def.ParseFromString(f.read())
+    return graph_def
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Export mask_rcnn_inception_v2_coco_2018_01_28.pb to ONNX")
+    parser.add_argument("--pb", default="../pb/mask_rcnn_inception_v2_coco_2018_01_28.pb")
+    parser.add_argument("--opset", type=int, default=18)
+    args = parser.parse_args()
+
+    graph_def = load_graph_def(args.pb)
+
+    model_proto, _ = tf2onnx.convert.from_graph_def(
+        graph_def,
+        input_names=["image_tensor:0"],
+        output_names=[
+            "num_detections:0",
+            "detection_boxes:0",
+            "detection_scores:0",
+            "detection_classes:0",
+            "detection_masks:0",
+        ],
+        opset=args.opset,
+    )
+    onnx.checker.check_model(model_proto)
+
+    stamp = datetime.datetime.now().strftime("%Y%b").lower()
+    onnx_path = "mask_rcnn_inception_v2_coco_2018_01_28_%s.onnx" % stamp
+    with open(onnx_path, "wb") as f:
+        f.write(model_proto.SerializeToString())
+    print("wrote", onnx_path)
+
+
+if __name__ == "__main__":
+    main()

mask_rcnn_inception_v2_coco_2018_01_28/demo.cpp

@@ -0,0 +1,126 @@
+#include <onnxruntime_cxx_api.h>
+#include <opencv2/imgproc.hpp>
+#include <opencv2/imgcodecs.hpp>
+#include <algorithm>
+#include <array>
+#include <cstdint>
+#include <iostream>
+#include <string>
+#include <vector>
+
+static std::string argVal(int argc, char** argv, const std::string& key, const std::string& def)
+{
+    for (int i = 1; i + 1 < argc; ++i)
+        if (key == argv[i]) return argv[i + 1];
+    return def;
+}
+
+int main(int argc, char** argv)
+{
+    std::string model  = argVal(argc, argv, "--model",  "mask_rcnn_inception_v2_coco_2018_01_28_2026jul.onnx");
+    std::string image  = argVal(argc, argv, "--image",  "example_outputs/input_image.png");
+    std::string output = argVal(argc, argv, "--output", "example_outputs/output_image.png");
+    float conf = std::stof(argVal(argc, argv, "--conf", "0.3"));
+
+    cv::Mat img = cv::imread(image);
+    if (img.empty())
+    {
+        std::cerr << "could not read image: " << image << std::endl;
+        return 1;
+    }
+
+    const int W = 800, H = 800;
+    cv::Mat rgb;
+    cv::cvtColor(img, rgb, cv::COLOR_BGR2RGB);
+    cv::resize(rgb, rgb, cv::Size(W, H));
+    if (!rgb.isContinuous()) rgb = rgb.clone();
+
+    Ort::Env env(ORT_LOGGING_LEVEL_WARNING, "demo");
+    Ort::SessionOptions so;
+    Ort::Session session(env, model.c_str(), so);
+    Ort::AllocatorWithDefaultOptions alloc;
+
+    auto in_name = session.GetInputNameAllocated(0, alloc);
+    const char* in_names[] = {in_name.get()};
+
+    size_t nout = session.GetOutputCount();
+    std::vector<Ort::AllocatedStringPtr> out_holders;
+    std::vector<std::string> out_strs;
+    for (size_t i = 0; i < nout; ++i)
+    {
+        out_holders.push_back(session.GetOutputNameAllocated(i, alloc));
+        out_strs.push_back(out_holders.back().get());
+    }
+    std::vector<const char*> out_names;
+    for (auto& s : out_strs) out_names.push_back(s.c_str());
+
+    std::array<int64_t, 4> shape = {1, H, W, 3};
+    auto mem = Ort::MemoryInfo::CreateCpu(OrtArenaAllocator, OrtMemTypeDefault);
+    Ort::Value input = Ort::Value::CreateTensor<uint8_t>(mem, rgb.data, (size_t)H * W * 3, shape.data(), shape.size());
+
+    auto outs = session.Run(Ort::RunOptions{nullptr}, in_names, &input, 1, out_names.data(), out_names.size());
+
+    float* boxes = nullptr;
+    float* scores = nullptr;
+    float* classes = nullptr;
+    float* numd = nullptr;
+    float* masks = nullptr;
+    for (size_t i = 0; i < nout; ++i)
+    {
+        float* p = outs[i].GetTensorMutableData<float>();
+        const std::string& n = out_strs[i];
+        if (n.find("detection_boxes") != std::string::npos) boxes = p;
+        else if (n.find("detection_scores") != std::string::npos) scores = p;
+        else if (n.find("detection_classes") != std::string::npos) classes = p;
+        else if (n.find("num_detections") != std::string::npos) numd = p;
+        else if (n.find("detection_masks") != std::string::npos) masks = p;
+    }
+    int nd = (int)numd[0];
+
+    int w = img.cols, h = img.rows;
+    cv::Mat out = img.clone();
+    std::vector<int> kept;
+    for (int i = 0; i < nd; ++i)
+        if (scores[i] >= conf) kept.push_back(i);
+
+    std::cout << "mask_rcnn_inception_v2_coco_2018_01_28 " << kept.size() << " detections" << std::endl;
+    for (int i : kept)
+    {
+        int cls = (int)classes[i] - 1;
+        float score = scores[i];
+        float y1 = boxes[i * 4 + 0], x1 = boxes[i * 4 + 1];
+        float y2 = boxes[i * 4 + 2], x2 = boxes[i * 4 + 3];
+        int px1 = std::max(0, (int)(x1 * w)), py1 = std::max(0, (int)(y1 * h));
+        int px2 = std::min(w, (int)(x2 * w)), py2 = std::min(h, (int)(y2 * h));
+
+        unsigned s = (unsigned)i * 2654435761u + 1u;
+        int col[3];
+        for (int c = 0; c < 3; ++c) { s = s * 1664525u + 1013904223u; col[c] = 80 + (int)((s >> 8) % 176u); }
+        cv::Scalar color(col[0], col[1], col[2]);
+
+        if (px2 > px1 && py2 > py1)
+        {
+            cv::Mat m15(15, 15, CV_32F, masks + (size_t)i * 225);
+            cv::Mat mr;
+            cv::resize(m15, mr, cv::Size(px2 - px1, py2 - py1));
+            cv::Mat roi = out(cv::Rect(px1, py1, px2 - px1, py2 - py1));
+            for (int y = 0; y < roi.rows; ++y)
+            {
+                cv::Vec3b* rp = roi.ptr<cv::Vec3b>(y);
+                const float* mp = mr.ptr<float>(y);
+                for (int x = 0; x < roi.cols; ++x)
+                    if (mp[x] > 0.5f)
+                        for (int c = 0; c < 3; ++c)
+                            rp[x][c] = (uchar)(0.5 * rp[x][c] + 0.5 * col[c]);
+            }
+        }
+        cv::rectangle(out, cv::Point(px1, py1), cv::Point(px2, py2), color, 2);
+        cv::putText(out, cv::format("%d:%.2f", cls, score), cv::Point(px1, py1 - 5),
+                    cv::FONT_HERSHEY_SIMPLEX, 0.5, color, 1);
+        std::cout << cls << " " << cv::format("%.3f %.3f %.3f %.3f %.3f", score, x1, y1, x2, y2) << std::endl;
+    }
+
+    cv::imwrite(output, out);
+    std::cout << "wrote " << output << std::endl;
+    return 0;
+}

mask_rcnn_inception_v2_coco_2018_01_28/demo.py

@@ -0,0 +1,59 @@
+import argparse
+import glob
+import os
+
+import cv2 as cv
+import numpy as np
+import onnxruntime as ort
+
+here = os.path.dirname(os.path.abspath(__file__))
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Mask-RCNN Inception v2 COCO (ONNX Runtime) detection + mask demo")
+    found = glob.glob(os.path.join(here, "*.onnx"))
+    parser.add_argument("--model", default=found[0] if found else None)
+    parser.add_argument("--image", default=os.path.join(here, "example_outputs", "input_image.png"))
+    parser.add_argument("--output", default=os.path.join(here, "example_outputs", "output_image.png"))
+    parser.add_argument("--conf", type=float, default=0.3)
+    args = parser.parse_args()
+
+    img = cv.imread(args.image)
+    if img is None:
+        raise SystemExit("could not read image: %s" % args.image)
+
+    sess = ort.InferenceSession(args.model, providers=["CPUExecutionProvider"])
+    rgb = cv.resize(cv.cvtColor(img, cv.COLOR_BGR2RGB), (800, 800))
+    res = sess.run(None, {sess.get_inputs()[0].name: rgb[None].astype(np.uint8)})
+    onames = [o.name for o in sess.get_outputs()]
+    boxes = res[[i for i, n in enumerate(onames) if "detection_boxes" in n][0]].reshape(-1, 4)
+    scores = res[[i for i, n in enumerate(onames) if "detection_scores" in n][0]].reshape(-1)
+    classes = res[[i for i, n in enumerate(onames) if "detection_classes" in n][0]].reshape(-1)
+    nd = int(res[[i for i, n in enumerate(onames) if "num_detections" in n][0]].reshape(-1)[0])
+    masks = res[[i for i, n in enumerate(onames) if "detection_masks" in n][0]].reshape(-1, 15, 15)
+
+    h, w = img.shape[:2]
+    out = img.copy()
+    kept = [i for i in range(nd) if scores[i] >= args.conf]
+    print("mask_rcnn_inception_v2_coco_2018_01_28", len(kept), "detections")
+    for i in kept:
+        cls = int(classes[i]) - 1
+        score = float(scores[i])
+        y1, x1, y2, x2 = boxes[i]
+        px1, py1 = max(0, int(x1 * w)), max(0, int(y1 * h))
+        px2, py2 = min(w, int(x2 * w)), min(h, int(y2 * h))
+        color = tuple(int(c) for c in np.random.default_rng(i).integers(80, 256, 3))
+        if px2 > px1 and py2 > py1:
+            m = cv.resize(masks[i], (px2 - px1, py2 - py1)) > 0.5
+            roi = out[py1:py2, px1:px2]
+            roi[m] = (0.5 * roi[m] + 0.5 * np.array(color)).astype(np.uint8)
+        cv.rectangle(out, (px1, py1), (px2, py2), color, 2)
+        cv.putText(out, "%d:%.2f" % (cls, score), (px1, py1 - 5), cv.FONT_HERSHEY_SIMPLEX, 0.5, color, 1)
+        print(cls, round(score, 3), round(float(x1), 3), round(float(y1), 3), round(float(x2), 3), round(float(y2), 3))
+
+    cv.imwrite(args.output, out)
+    print("wrote", args.output)
+
+
+if __name__ == "__main__":
+    main()

mask_rcnn_inception_v2_coco_2018_01_28/mask_rcnn_inception_v2_coco_2018_01_28_2026jul.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27cdd89df33ab94f61fd278350f66996efec3437937b5679a311ba55e386690f
+size 66728941

opencv_face_detector_uint8/LICENSE

@@ -0,0 +1,203 @@
+Copyright (c) OpenCV team and the opencv_3rdparty contributors.  All rights reserved.
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright (c) OpenCV team and the opencv_3rdparty contributors.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+limitations under the License.

opencv_face_detector_uint8/README.md

@@ -0,0 +1,71 @@
+# OpenCV SSD Face Detector (UINT8)
+
+Single-shot face detection with the OpenCV SSD ResNet-10 network. The model ships in the
+OpenCV project as a quantized frozen TensorFlow graph (`opencv_face_detector_uint8.pb`) and is
+converted here to ONNX for use with OpenCV's DNN module and onnxruntime. Only the backbone is
+exported — PriorBox generation, the confidence softmax, variance decode, score threshold and NMS
+are run in host code (see `demo.py` / `demo.cpp`).
+
+## Model Details
+- **Architecture**: SSD with a ResNet-10 backbone (face detector)
+- **Input**: BGR image, 300×300, mean-subtracted by `[104, 177, 123]` (no scaling, no RGB swap),
+  NHWC layout (`data:0`, shape `[1, 300, 300, 3]`)
+- **Output**: `mbox_loc` (`[1, 35568]`, box regressions) and `mbox_conf_flatten`
+  (`[1, 17784]`, 2-class face/background logits); PriorBox decode + softmax + NMS are done in the demo
+- **Framework**: ONNX (converted from the TensorFlow frozen graph — uint8 weights with the
+  `Dequantize` nodes folded to float `Const` — via tf2onnx, opset 18)
+- **Original weights**: https://github.com/opencv/opencv_3rdparty/raw/8033c2bc31b3256f0d461c919ecc01c2428ca03b/opencv_face_detector_uint8.pb
+
+The 6 SSD prior layers (min/max size, aspect ratios, step, feature-map size), the variances
+`[0.1, 0.1, 0.2, 0.2]`, the default confidence threshold `0.4` and the NMS IoU `0.3` are all
+defined in the demo scripts.
+
+## Usage
+
+### Python
+```bash
+python demo.py --model opencv_face_detector_uint8_2026jul.onnx --image example_outputs/input_image.png --output example_outputs/output_image.png --conf 0.4
+```
+
+Or import directly:
+```python
+import cv2
+
+net = cv2.dnn.readNet("opencv_face_detector_uint8_2026jul.onnx")
+# see demo.py for the full PriorBox decode + softmax + NMS pipeline
+```
+
+### C++
+The C++ demo runs inference with ONNX Runtime (C++ API) and uses OpenCV only for image I/O.
+Install ONNX Runtime (C++) from https://github.com/microsoft/onnxruntime/releases — this build
+uses `onnxruntime-linux-x64-1.25.0` — and adjust the ONNX Runtime and OpenCV paths to your setup:
+```bash
+ORT=/path/to/onnxruntime-linux-x64-1.25.0    # ONNX Runtime release dir (contains include/ and lib/)
+OCV=/path/to/opencv                          # OpenCV source tree
+OCVBUILD=/path/to/opencv/build               # OpenCV build directory (generated headers + libs)
+g++ -std=c++17 demo.cpp -o demo \
+  -I$ORT/include \
+  -I$OCV/include \
+  -I$OCV/modules/core/include \
+  -I$OCV/modules/imgproc/include \
+  -I$OCV/modules/imgcodecs/include \
+  -I$OCVBUILD \
+  -L$ORT/lib -Wl,-rpath,$ORT/lib -lonnxruntime \
+  -L$OCVBUILD/lib -Wl,-rpath,$OCVBUILD/lib -lopencv_imgcodecs -lopencv_imgproc -lopencv_core
+./demo --model opencv_face_detector_uint8_2026jul.onnx --image example_outputs/input_image.png --output example_outputs/output_image.png
+```
+
+## Conversion
+The ONNX model was exported from the frozen TensorFlow graph with tf2onnx (opset 18) via
+[convert_to_onnx.py](./convert_to_onnx.py). The `.pb` stores its weights behind `Dequantize`
+nodes, so the script first folds every `Dequantize` node to a float `Const` before conversion.
+Inputs `data:0`, outputs `mbox_loc:0` and `mbox_conf_flatten:0`, input shape overridden to
+`[1, 300, 300, 3]`. Requires `tensorflow`, `tf2onnx`, and `onnx`.
+
+```bash
+python convert_to_onnx.py --pb ../pb/opencv_face_detector_uint8.pb
+```
+
+## License
+See [LICENSE](./LICENSE) — this is the OpenCV face detector distributed via `opencv_3rdparty`
+under the Apache License 2.0.

opencv_face_detector_uint8/convert_to_onnx.py

@@ -0,0 +1,73 @@
+import argparse
+import datetime
+
+import numpy as np
+import tensorflow as tf
+import tf2onnx
+import onnx
+from tensorflow.python.framework import graph_util, tensor_util
+
+
+def dequantize(graph_def, outputs):
+    dmin = {n.name: n.input[1] for n in graph_def.node if n.op == "Dequantize"}
+    folded = {}
+    with tf.Graph().as_default() as g:
+        tf.import_graph_def(graph_def, name="")
+        deq = list(dmin)
+        with tf.compat.v1.Session(graph=g) as sess:
+            for name in deq:
+                v = np.asarray(sess.run(g.get_tensor_by_name(name + ":0")), np.float32)
+                if not np.isfinite(v).all():
+                    mn = np.float32(sess.run(g.get_tensor_by_name(dmin[name] + ":0")))
+                    v = np.full(v.shape, mn, np.float32)
+                folded[name] = v
+    new = tf.compat.v1.GraphDef()
+    for n in graph_def.node:
+        if n.op == "Dequantize":
+            v = folded[n.name]
+            c = new.node.add()
+            c.op = "Const"
+            c.name = n.name
+            c.attr["dtype"].type = tf.float32.as_datatype_enum
+            c.attr["value"].tensor.CopyFrom(tensor_util.make_tensor_proto(v, tf.float32, v.shape))
+        else:
+            new.node.add().CopyFrom(n)
+    return graph_util.extract_sub_graph(new, [o.split(":")[0] for o in outputs])
+
+
+def load_graph_def(pb_path):
+    with tf.io.gfile.GFile(pb_path, "rb") as f:
+        graph_def = tf.compat.v1.GraphDef()
+        graph_def.ParseFromString(f.read())
+    return graph_def
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Export opencv_face_detector_uint8.pb (backbone) to ONNX")
+    parser.add_argument("--pb", default="../pb/opencv_face_detector_uint8.pb")
+    parser.add_argument("--opset", type=int, default=18)
+    args = parser.parse_args()
+
+    output_names = ["mbox_loc:0", "mbox_conf_flatten:0"]
+
+    graph_def = load_graph_def(args.pb)
+    graph_def = dequantize(graph_def, output_names)
+
+    model_proto, _ = tf2onnx.convert.from_graph_def(
+        graph_def,
+        input_names=["data:0"],
+        output_names=output_names,
+        opset=args.opset,
+        shape_override={"data:0": [1, 300, 300, 3]},
+    )
+    onnx.checker.check_model(model_proto)
+
+    stamp = datetime.datetime.now().strftime("%Y%b").lower()
+    onnx_path = "opencv_face_detector_uint8_%s.onnx" % stamp
+    with open(onnx_path, "wb") as f:
+        f.write(model_proto.SerializeToString())
+    print("wrote", onnx_path)
+
+
+if __name__ == "__main__":
+    main()

opencv_face_detector_uint8/demo.cpp

@@ -0,0 +1,165 @@
+#include <onnxruntime_cxx_api.h>
+#include <opencv2/imgproc.hpp>
+#include <opencv2/imgcodecs.hpp>
+#include <algorithm>
+#include <array>
+#include <cmath>
+#include <iostream>
+#include <string>
+#include <vector>
+
+using namespace cv;
+
+static std::string argVal(int argc, char** argv, const std::string& key, const std::string& def)
+{
+    for (int i = 1; i + 1 < argc; ++i)
+        if (key == argv[i]) return argv[i + 1];
+    return def;
+}
+
+struct Layer { float mn, mx; std::vector<int> ars; int step, fm; };
+
+int main(int argc, char** argv)
+{
+    std::string model  = argVal(argc, argv, "--model",  "opencv_face_detector_uint8_2026jul.onnx");
+    std::string image  = argVal(argc, argv, "--image",  "example_outputs/input_image.png");
+    std::string output = argVal(argc, argv, "--output", "example_outputs/output_image.png");
+    float thr = std::stof(argVal(argc, argv, "--conf", "0.4"));
+
+    const int sz = 300;
+    Mat img = imread(image);
+    if (img.empty())
+    {
+        std::cerr << "could not read image: " << image << std::endl;
+        return 1;
+    }
+
+    Mat inp;
+    resize(img, inp, Size(sz, sz));
+    inp.convertTo(inp, CV_32F);
+    subtract(inp, Scalar(104, 177, 123), inp);
+    if (!inp.isContinuous()) inp = inp.clone();
+
+    Ort::Env env(ORT_LOGGING_LEVEL_WARNING, "demo");
+    Ort::SessionOptions so;
+    Ort::Session session(env, model.c_str(), so);
+    Ort::AllocatorWithDefaultOptions alloc;
+
+    auto in_name = session.GetInputNameAllocated(0, alloc);
+    const char* in_names[] = {in_name.get()};
+
+    size_t nout = session.GetOutputCount();
+    std::vector<Ort::AllocatedStringPtr> out_holders;
+    std::vector<std::string> out_str;
+    for (size_t i = 0; i < nout; ++i)
+    {
+        out_holders.push_back(session.GetOutputNameAllocated(i, alloc));
+        out_str.push_back(out_holders.back().get());
+    }
+    std::vector<const char*> out_names;
+    for (auto& s : out_str) out_names.push_back(s.c_str());
+
+    std::array<int64_t, 4> shape = {1, sz, sz, 3};
+    auto mem = Ort::MemoryInfo::CreateCpu(OrtArenaAllocator, OrtMemTypeDefault);
+    Ort::Value input = Ort::Value::CreateTensor<float>(mem, (float*)inp.data, (size_t)sz * sz * 3, shape.data(), shape.size());
+
+    auto outs = session.Run(Ort::RunOptions{nullptr}, in_names, &input, 1, out_names.data(), out_names.size());
+
+    const float* loc = nullptr;
+    const float* conf = nullptr;
+    for (size_t i = 0; i < outs.size(); ++i)
+    {
+        auto os = outs[i].GetTensorTypeAndShapeInfo().GetShape();
+        size_t tot = 1;
+        for (auto d : os) tot *= (size_t)d;
+        const float* p = outs[i].GetTensorMutableData<float>();
+        if (tot == 35568) loc = p;
+        else if (tot == 17784) conf = p;
+    }
+
+    std::vector<Layer> layers = {
+        {30, 60, {2}, 8, 38},
+        {60, 111, {2, 3}, 16, 19},
+        {111, 162, {2, 3}, 32, 10},
+        {162, 213, {2, 3}, 64, 5},
+        {213, 264, {2}, 100, 5},
+        {264, 315, {2}, 300, 5},
+    };
+    std::vector<Vec4f> priors;
+    for (const Layer& L : layers)
+    {
+        std::vector<float> ratios = {1.0f};
+        for (int a : L.ars) { ratios.push_back((float)a); ratios.push_back(1.0f / a); }
+        for (int y = 0; y < L.fm; ++y)
+            for (int x = 0; x < L.fm; ++x)
+            {
+                float cx = (x + 0.5f) * L.step;
+                float cy = (y + 0.5f) * L.step;
+                std::vector<Vec2f> boxes = {{L.mn, L.mn}, {std::sqrt(L.mn * L.mx), std::sqrt(L.mn * L.mx)}};
+                for (size_t k = 1; k < ratios.size(); ++k)
+                {
+                    float a = ratios[k];
+                    boxes.push_back({L.mn * std::sqrt(a), L.mn / std::sqrt(a)});
+                }
+                for (const Vec2f& b : boxes)
+                    priors.push_back({cx, cy, b[0], b[1]});
+            }
+    }
+
+    const float var[4] = {0.1f, 0.1f, 0.2f, 0.2f};
+    int n = (int)priors.size();
+    std::vector<Rect2f> boxes;
+    std::vector<float> scores;
+    for (int i = 0; i < n; ++i)
+    {
+        float c0 = conf[i * 2], c1 = conf[i * 2 + 1];
+        float m = std::max(c0, c1);
+        float e0 = std::exp(c0 - m), e1 = std::exp(c1 - m);
+        float s = e1 / (e0 + e1);
+        if (s <= thr) continue;
+        float pcx = priors[i][0] / sz, pcy = priors[i][1] / sz;
+        float pw = priors[i][2] / sz, ph = priors[i][3] / sz;
+        float cx = pcx + loc[i * 4] * var[0] * pw;
+        float cy = pcy + loc[i * 4 + 1] * var[1] * ph;
+        float bw = pw * std::exp(loc[i * 4 + 2] * var[2]);
+        float bh = ph * std::exp(loc[i * 4 + 3] * var[3]);
+        boxes.push_back(Rect2f(cx - bw / 2, cy - bh / 2, bw, bh));
+        scores.push_back(s);
+    }
+
+    std::vector<int> order(scores.size());
+    for (size_t i = 0; i < order.size(); ++i) order[i] = (int)i;
+    std::sort(order.begin(), order.end(), [&](int a, int b){ return scores[a] > scores[b]; });
+    std::vector<char> removed(order.size(), 0);
+    std::vector<int> pick;
+    for (size_t oi = 0; oi < order.size(); ++oi)
+    {
+        if (removed[oi]) continue;
+        int i = order[oi];
+        pick.push_back(i);
+        for (size_t oj = oi + 1; oj < order.size(); ++oj)
+        {
+            if (removed[oj]) continue;
+            int j = order[oj];
+            const Rect2f& a = boxes[i];
+            const Rect2f& b = boxes[j];
+            float xx1 = std::max(a.x, b.x), yy1 = std::max(a.y, b.y);
+            float xx2 = std::min(a.x + a.width, b.x + b.width);
+            float yy2 = std::min(a.y + a.height, b.y + b.height);
+            float inter = std::max(0.f, xx2 - xx1) * std::max(0.f, yy2 - yy1);
+            float iou = inter / (a.area() + b.area() - inter + 1e-9f);
+            if (iou > 0.3f) removed[oj] = 1;
+        }
+    }
+
+    int W = img.cols, H = img.rows;
+    for (int i : pick)
+    {
+        const Rect2f& b = boxes[i];
+        rectangle(img, Point(int(b.x * W), int(b.y * H)),
+                  Point(int((b.x + b.width) * W), int((b.y + b.height) * H)), Scalar(0, 255, 0), 2);
+    }
+    imwrite(output, img);
+    std::cout << "opencv_face_detector_uint8 " << pick.size() << " faces" << std::endl;
+    return 0;
+}

opencv_face_detector_uint8/demo.py

@@ -0,0 +1,109 @@
+import argparse
+import glob
+import os
+
+import cv2 as cv
+import numpy as np
+import onnxruntime as ort
+
+here = os.path.dirname(os.path.abspath(__file__))
+
+sz = 300
+layers = [
+    (30, 60, [2], 8, 38),
+    (60, 111, [2, 3], 16, 19),
+    (111, 162, [2, 3], 32, 10),
+    (162, 213, [2, 3], 64, 5),
+    (213, 264, [2], 100, 5),
+    (264, 315, [2], 300, 5),
+]
+var = [0.1, 0.1, 0.2, 0.2]
+
+
+def build_priors():
+    p = []
+    for mn, mx, ars, step, fm in layers:
+        ratios = [1.0]
+        for a in ars:
+            ratios += [a, 1.0 / a]
+        for y in range(fm):
+            for x in range(fm):
+                cx = (x + 0.5) * step
+                cy = (y + 0.5) * step
+                boxes = [(mn, mn), ((mn * mx) ** 0.5, (mn * mx) ** 0.5)]
+                for a in ratios[1:]:
+                    boxes.append((mn * a ** 0.5, mn / a ** 0.5))
+                for bw, bh in boxes:
+                    p.append([cx, cy, bw, bh])
+    return np.array(p, np.float32)
+
+
+def default_model():
+    files = [f for f in glob.glob(os.path.join(here, "*.onnx")) if "known_good" not in os.path.basename(f)]
+    return files[0] if files else os.path.join(here, "opencv_face_detector_uint8.onnx")
+
+
+def main():
+    parser = argparse.ArgumentParser(description="OpenCV SSD face detector (ONNX) demo")
+    parser.add_argument("--model", default=default_model())
+    parser.add_argument("--image", default=os.path.join(here, "example_outputs", "input_image.png"))
+    parser.add_argument("--output", default=os.path.join(here, "example_outputs", "output_image.png"))
+    parser.add_argument("--conf", type=float, default=0.4)
+    args = parser.parse_args()
+
+    img = cv.imread(args.image)
+    if img is None:
+        raise SystemExit("could not read image: %s" % args.image)
+
+    inp = cv.resize(img, (sz, sz)).astype(np.float32) - np.array([104.0, 177.0, 123.0], np.float32)
+
+    sess = ort.InferenceSession(args.model, providers=["CPUExecutionProvider"])
+    res = sess.run(None, {sess.get_inputs()[0].name: inp[None]})
+    onames = [o.name for o in sess.get_outputs()]
+    loc = res[[i for i, n in enumerate(onames) if "mbox_loc" in n][0]].reshape(-1, 4)
+    conf = res[[i for i, n in enumerate(onames) if "mbox_conf" in n][0]].reshape(-1, 2)
+
+    priors = build_priors()
+    pcx = priors[:, 0] / sz
+    pcy = priors[:, 1] / sz
+    pw = priors[:, 2] / sz
+    ph = priors[:, 3] / sz
+
+    e = np.exp(conf - conf.max(1, keepdims=True))
+    sm = e / e.sum(1, keepdims=True)
+    scores = sm[:, 1]
+
+    cx = pcx + loc[:, 0] * var[0] * pw
+    cy = pcy + loc[:, 1] * var[1] * ph
+    bw = pw * np.exp(loc[:, 2] * var[2])
+    bh = ph * np.exp(loc[:, 3] * var[3])
+    boxes = np.stack([cx - bw / 2, cy - bh / 2, cx + bw / 2, cy + bh / 2], 1)
+
+    keep = scores > args.conf
+    boxes = boxes[keep]
+    scores = scores[keep]
+    order = scores.argsort()[::-1]
+    pick = []
+    while order.size:
+        i = order[0]
+        pick.append(i)
+        xx1 = np.maximum(boxes[i, 0], boxes[order[1:], 0])
+        yy1 = np.maximum(boxes[i, 1], boxes[order[1:], 1])
+        xx2 = np.minimum(boxes[i, 2], boxes[order[1:], 2])
+        yy2 = np.minimum(boxes[i, 3], boxes[order[1:], 3])
+        inter = np.maximum(0, xx2 - xx1) * np.maximum(0, yy2 - yy1)
+        ai = (boxes[i, 2] - boxes[i, 0]) * (boxes[i, 3] - boxes[i, 1])
+        aj = (boxes[order[1:], 2] - boxes[order[1:], 0]) * (boxes[order[1:], 3] - boxes[order[1:], 1])
+        iou = inter / (ai + aj - inter + 1e-9)
+        order = order[1:][iou <= 0.3]
+
+    h, w = img.shape[:2]
+    for i in pick:
+        x1, y1, x2, y2 = boxes[i]
+        cv.rectangle(img, (int(x1 * w), int(y1 * h)), (int(x2 * w), int(y2 * h)), (0, 255, 0), 2)
+    cv.imwrite(args.output, img)
+    print("opencv_face_detector_uint8", len(pick), "faces")
+
+
+if __name__ == "__main__":
+    main()

opencv_face_detector_uint8/opencv_face_detector_uint8_2026jul.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5b1efe9c4e792a010ac36248d92b64c3d94f9bdec764951d1e8684d919b1e40
+size 10671719

ssd_inception_v2_coco_2017_11_17/LICENSE

@@ -0,0 +1,212 @@
+Copyright 2022 Google LLC. All rights reserved.
+
+All files in the following folders:
+/community
+/official
+/orbit
+/research
+/tensorflow_models
+
+Are licensed as follows:
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2016, The Authors.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

ssd_inception_v2_coco_2017_11_17/README.md

@@ -0,0 +1,52 @@
+# SSD Inception v2 COCO
+
+Object detection with a Single Shot MultiBox Detector (SSD) built on an Inception v2 backbone,
+trained on the COCO dataset. The model was originally distributed as a frozen TensorFlow graph
+(`ssd_inception_v2_coco_2017_11_17.pb`) and converted to ONNX for use with OpenCV's DNN module.
+
+## Model Details
+- **Architecture**: SSD (Single Shot MultiBox Detector) with Inception v2 backbone
+- **Input**: RGB image, 300×300, raw uint8, NHWC layout (`image_tensor:0`, shape `[1, 300, 300, 3]`)
+- **Output**: `detection_boxes:0` (normalized `ymin, xmin, ymax, xmax`), `detection_scores:0`, `detection_classes:0` (COCO class ids), `num_detections:0`
+- **Framework**: ONNX (converted from the TensorFlow frozen graph via tf2onnx, opset 18)
+- **Original weights**: http://download.tensorflow.org/models/object_detection/ssd_inception_v2_coco_2017_11_17.tar.gz
+
+## Usage
+
+### Python
+```bash
+python demo.py --model ssd_inception_v2_coco_2017_11_17_2026jul.onnx --image example_outputs/input_image.png --output example_outputs/output_image.png --conf 0.3
+```
+
+### C++
+The C++ demo runs inference with ONNX Runtime (C++ API) and uses OpenCV only for image I/O.
+Install ONNX Runtime (C++) from https://github.com/microsoft/onnxruntime/releases — this build
+uses `onnxruntime-linux-x64-1.25.0` — and adjust the ONNX Runtime and OpenCV paths to your setup:
+```bash
+ORT=/path/to/onnxruntime-linux-x64-1.25.0    # ONNX Runtime release dir (contains include/ and lib/)
+OCV=/path/to/opencv                          # OpenCV source tree
+OCVBUILD=/path/to/opencv/build               # OpenCV build directory (generated headers + libs)
+g++ -std=c++17 demo.cpp -o demo \
+  -I$ORT/include \
+  -I$OCV/include \
+  -I$OCV/modules/core/include \
+  -I$OCV/modules/imgproc/include \
+  -I$OCV/modules/imgcodecs/include \
+  -I$OCVBUILD \
+  -L$ORT/lib -Wl,-rpath,$ORT/lib -lonnxruntime \
+  -L$OCVBUILD/lib -Wl,-rpath,$OCVBUILD/lib -lopencv_imgcodecs -lopencv_imgproc -lopencv_core
+./demo --model ssd_inception_v2_coco_2017_11_17_2026jul.onnx --image example_outputs/input_image.png --output example_outputs/output_image.png
+```
+
+## Conversion
+The ONNX model was exported from the frozen TensorFlow graph with tf2onnx (opset 18)
+via [convert_to_onnx.py](./convert_to_onnx.py) — input `image_tensor:0`, outputs
+`detection_boxes:0`, `detection_scores:0`, `detection_classes:0`, `num_detections:0`.
+Requires `tensorflow`, `tf2onnx`, and `onnx`.
+
+```bash
+python convert_to_onnx.py --pb ../pb/ssd_inception_v2_coco_2017_11_17.pb
+```
+
+## License
+See [LICENSE](./LICENSE) — the model is released by the TensorFlow Authors under the Apache License 2.0.

ssd_inception_v2_coco_2017_11_17/convert_to_onnx.py

@@ -0,0 +1,40 @@
+import argparse
+import datetime
+
+import onnx
+import tensorflow as tf
+import tf2onnx
+
+
+def load_graph_def(pb_path):
+    with tf.io.gfile.GFile(pb_path, "rb") as f:
+        graph_def = tf.compat.v1.GraphDef()
+        graph_def.ParseFromString(f.read())
+    return graph_def
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Export ssd_inception_v2_coco_2017_11_17.pb to ONNX")
+    parser.add_argument("--pb", default="../pb/ssd_inception_v2_coco_2017_11_17.pb")
+    parser.add_argument("--opset", type=int, default=18)
+    args = parser.parse_args()
+
+    graph_def = load_graph_def(args.pb)
+
+    model_proto, _ = tf2onnx.convert.from_graph_def(
+        graph_def,
+        input_names=["image_tensor:0"],
+        output_names=["detection_boxes:0", "detection_scores:0", "detection_classes:0", "num_detections:0"],
+        opset=args.opset,
+    )
+    onnx.checker.check_model(model_proto)
+
+    stamp = datetime.datetime.now().strftime("%Y%b").lower()
+    onnx_path = "ssd_inception_v2_coco_2017_11_17_%s.onnx" % stamp
+    with open(onnx_path, "wb") as f:
+        f.write(model_proto.SerializeToString())
+    print("wrote", onnx_path)
+
+
+if __name__ == "__main__":
+    main()

ssd_inception_v2_coco_2017_11_17/demo.cpp

@@ -0,0 +1,98 @@
+#include <onnxruntime_cxx_api.h>
+#include <opencv2/imgproc.hpp>
+#include <opencv2/imgcodecs.hpp>
+#include <array>
+#include <cstdint>
+#include <iostream>
+#include <string>
+#include <vector>
+
+using namespace cv;
+
+static std::string argVal(int argc, char** argv, const std::string& key, const std::string& def)
+{
+    for (int i = 1; i + 1 < argc; ++i)
+        if (key == argv[i]) return argv[i + 1];
+    return def;
+}
+
+int main(int argc, char** argv)
+{
+    std::string model  = argVal(argc, argv, "--model",  "ssd_inception_v2_coco_2017_11_17_2026jul.onnx");
+    std::string image  = argVal(argc, argv, "--image",  "example_outputs/input_image.png");
+    std::string output = argVal(argc, argv, "--output", "example_outputs/output_image.png");
+    float conf = std::stof(argVal(argc, argv, "--conf", "0.3"));
+
+    Mat img = imread(image);
+    if (img.empty())
+    {
+        std::cerr << "could not read image: " << image << std::endl;
+        return 1;
+    }
+
+    Mat rgb;
+    cvtColor(img, rgb, COLOR_BGR2RGB);
+    resize(rgb, rgb, Size(300, 300));
+    if (!rgb.isContinuous()) rgb = rgb.clone();
+
+    Ort::Env env(ORT_LOGGING_LEVEL_WARNING, "demo");
+    Ort::SessionOptions so;
+    Ort::Session session(env, model.c_str(), so);
+    Ort::AllocatorWithDefaultOptions alloc;
+
+    auto in_name = session.GetInputNameAllocated(0, alloc);
+    const char* in_names[] = {in_name.get()};
+
+    size_t out_count = session.GetOutputCount();
+    std::vector<Ort::AllocatedStringPtr> out_holders;
+    std::vector<std::string> out_str;
+    std::vector<const char*> out_names;
+    for (size_t i = 0; i < out_count; ++i)
+    {
+        out_holders.push_back(session.GetOutputNameAllocated(i, alloc));
+        out_str.push_back(out_holders.back().get());
+        out_names.push_back(out_str.back().c_str());
+    }
+
+    std::array<int64_t, 4> shape = {1, 300, 300, 3};
+    auto mem = Ort::MemoryInfo::CreateCpu(OrtArenaAllocator, OrtMemTypeDefault);
+    Ort::Value input = Ort::Value::CreateTensor<uint8_t>(mem, rgb.data, 300 * 300 * 3, shape.data(), shape.size());
+
+    auto outs = session.Run(Ort::RunOptions{nullptr}, in_names, &input, 1, out_names.data(), out_names.size());
+
+    const float *boxes = 0, *scores = 0, *classes = 0, *num = 0;
+    for (size_t i = 0; i < out_count; ++i)
+    {
+        const std::string& n = out_str[i];
+        if (n.find("detection_boxes") != std::string::npos) boxes = outs[i].GetTensorMutableData<float>();
+        else if (n.find("detection_scores") != std::string::npos) scores = outs[i].GetTensorMutableData<float>();
+        else if (n.find("detection_classes") != std::string::npos) classes = outs[i].GetTensorMutableData<float>();
+        else if (n.find("num_detections") != std::string::npos) num = outs[i].GetTensorMutableData<float>();
+    }
+    if (!boxes || !scores || !classes || !num)
+    {
+        std::cerr << "missing expected output tensors" << std::endl;
+        return 1;
+    }
+
+    int nd = (int)num[0];
+    int h = img.rows, w = img.cols;
+    Mat out = img.clone();
+    std::vector<std::string> lines;
+    for (int k = 0; k < nd; ++k)
+    {
+        if (scores[k] < conf) continue;
+        float ymin = boxes[k * 4 + 0], xmin = boxes[k * 4 + 1];
+        float ymax = boxes[k * 4 + 2], xmax = boxes[k * 4 + 3];
+        int cls = (int)classes[k];
+        rectangle(out, Point(int(xmin * w), int(ymin * h)), Point(int(xmax * w), int(ymax * h)), Scalar(0, 255, 0), 2);
+        putText(out, format("%d:%.2f", cls, scores[k]), Point(int(xmin * w), int(ymin * h) - 5),
+                FONT_HERSHEY_SIMPLEX, 0.5, Scalar(0, 255, 0), 1);
+        lines.push_back(format("%d %.3f %.3f %.3f %.3f %.3f", cls, scores[k], xmin, ymin, xmax, ymax));
+    }
+
+    imwrite(output, out);
+    std::cout << "ssd_inception_v2_coco_2017_11_17 " << lines.size() << " detections" << std::endl;
+    for (const auto& l : lines) std::cout << l << std::endl;
+    return 0;
+}

ssd_inception_v2_coco_2017_11_17/demo.py

@@ -0,0 +1,52 @@
+import argparse
+import glob
+import os
+
+import cv2 as cv
+import numpy as np
+import onnxruntime as ort
+
+here = os.path.dirname(os.path.abspath(__file__))
+
+
+def main():
+    parser = argparse.ArgumentParser(description="SSD Inception v2 COCO (ONNX) object detection demo")
+    parser.add_argument("--model", default=(glob.glob(os.path.join(here, "*.onnx")) or [""])[0])
+    parser.add_argument("--image", default=os.path.join(here, "example_outputs", "input_image.png"))
+    parser.add_argument("--output", default=os.path.join(here, "example_outputs", "output_image.png"))
+    parser.add_argument("--conf", type=float, default=0.3)
+    args = parser.parse_args()
+
+    img = cv.imread(args.image)
+    if img is None:
+        raise SystemExit("could not read image: %s" % args.image)
+
+    rgb = cv.resize(cv.cvtColor(img, cv.COLOR_BGR2RGB), (300, 300))
+
+    sess = ort.InferenceSession(args.model, providers=["CPUExecutionProvider"])
+    res = sess.run(None, {sess.get_inputs()[0].name: rgb[None].astype(np.uint8)})
+    onames = [o.name for o in sess.get_outputs()]
+    boxes = res[[i for i, n in enumerate(onames) if "detection_boxes" in n][0]].reshape(-1, 4)
+    scores = res[[i for i, n in enumerate(onames) if "detection_scores" in n][0]].reshape(-1)
+    classes = res[[i for i, n in enumerate(onames) if "detection_classes" in n][0]].reshape(-1)
+    nd = int(res[[i for i, n in enumerate(onames) if "num_detections" in n][0]].reshape(-1)[0])
+
+    h, w = img.shape[:2]
+    out = img.copy()
+    kept = []
+    for k in range(nd):
+        if scores[k] < args.conf:
+            continue
+        ymin, xmin, ymax, xmax = boxes[k]
+        kept.append((int(classes[k]), float(scores[k]), float(xmin), float(ymin), float(xmax), float(ymax)))
+        cv.rectangle(out, (int(xmin * w), int(ymin * h)), (int(xmax * w), int(ymax * h)), (0, 255, 0), 2)
+        cv.putText(out, "%d:%.2f" % (int(classes[k]), scores[k]), (int(xmin * w), int(ymin * h) - 5), cv.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 0), 1)
+
+    cv.imwrite(args.output, out)
+    print("ssd_inception_v2_coco_2017_11_17", len(kept), "detections")
+    for c, s, xmin, ymin, xmax, ymax in kept:
+        print(c, round(s, 3), round(xmin, 3), round(ymin, 3), round(xmax, 3), round(ymax, 3))
+
+
+if __name__ == "__main__":
+    main()

ssd_inception_v2_coco_2017_11_17/ssd_inception_v2_coco_2017_11_17_2026jul.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5834f7214c7b9d80a9f708d23a3539d93e632fb346aeb5b1d013e63443354009
+size 102207336

ssd_mobilenet_v1_ppn_coco/LICENSE

@@ -0,0 +1,203 @@
+Copyright 2015 The TensorFlow Authors.  All rights reserved.
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2015, The TensorFlow Authors.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+limitations under the License.