update

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Omni Instrument Inc.
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

README.md

@@ -1,56 +1,209 @@
----
-language:
-- en
-license: mit
-pretty_name: omnislamproject
-task_categories:
-- robotics
-- depth-estimation
-- keypoint-detection
-configs:
-- config_name: default
-  data_files:
-  - split: stereo
-    path: data/stereo-*
-  - split: stereoinertial
-    path: data/stereoinertial-*
-  - split: vio
-    path: data/vio-*
-dataset_info:
-  features:
-  - name: image_left
-    dtype: image
-  - name: image_right
-    dtype: image
-  - name: timestamp
-    dtype: float64
-  - name: gyro
-    list: float32
-    length: 3
-  - name: accel
-    list: float32
-    length: 3
-  - name: sync_dt
-    list: float32
-    length: 2
-  - name: position
-    list: float32
-    length: 3
-  - name: orientation
-    list: float32
-    length: 4
-  splits:
-  - name: stereo
-    num_bytes: 381314848
-    num_examples: 100
-  - name: stereoinertial
-    num_bytes: 381134104
-    num_examples: 100
-  - name: vio
-    num_bytes: 370323077
-    num_examples: 100
-  download_size: 1132892975
-  dataset_size: 1132772029
----
-
-# Omni Instrument SLAM Project Dataset 
+---
+language:
+- en
+license: mit
+license_link: LICENSE
+pretty_name: omnislamproject
+task_categories:
+- robotics
+- depth-estimation
+- keypoint-detection
+configs:
+- config_name: default
+  data_files:
+  - split: stereo
+    path: data/stereo-*
+  - split: stereoinertial
+    path: data/stereoinertial-*
+  - split: vio
+    path: data/vio-*
+dataset_info:
+  features:
+  - name: image_left
+    dtype: image
+  - name: image_right
+    dtype: image
+  - name: timestamp
+    dtype: float64
+  - name: gyro
+    list: float32
+    length: 3
+  - name: accel
+    list: float32
+    length: 3
+  - name: sync_dt
+    list: float32
+    length: 2
+  - name: position
+    list: float32
+    length: 3
+  - name: orientation
+    list: float32
+    length: 4
+  splits:
+  - name: stereo
+    num_bytes: 381314848
+    num_examples: 100
+  - name: stereoinertial
+    num_bytes: 381134104
+    num_examples: 100
+  - name: vio
+    num_bytes: 370323077
+    num_examples: 100
+  download_size: 1132892975
+  dataset_size: 1132772029
+---
+# Omni Instrument SLAM Project Dataset
+
+The Omni Instrument SLAM Project Dataset is a compact robotics dataset designed for evaluating stereo, visual-inertial, and visual-inertial odometry (VIO) pipelines.
+
+It provides:
+- [x] Stereo Image Pairs
+- [x] Inertial measurements (IMU)
+- [x] Ground-truth 6DoF pose (for VIO)
+- [x] Raw ROS 1 and ROS 2 recordings
+
+## Overview
+
+The dataset is structured into three splits:
+
+| Split | Description |
+| --- | --- |
+| `stereo` | Stereo-only (IMU stationary) |
+| `stereoinertial` | Stereo + IMU |
+| `vio` | Stereo + IMU + ground-truth pose |
+
+All splits share the same schema, enabling consistent downstream pipelines.
+
+## Data Collection Protocol
+
+1. Stereo (Calibration - Static Sensor)
+   - Robot body stationary
+   - Camera and IMU fixed
+   - AprilTag calibration grid moves in front of the camera
+
+   Purpose:
+   - Stereo camera calibration
+   - Intrinsics/extrinsics estimation
+
+2. Stereo-Inertial (Calibration - Moving Sensor)
+   - Robot body moves
+   - Camera and IMU move together
+   - AprilTag calibration grid stationary
+
+   Purpose:
+   - Camera-IMU extrinsic calibration
+   - Temporal synchronization validation
+   - Motion-consistent calibration
+
+3. VIO (Operational SLAM Sequence)
+   - Robot moves freely in the environment
+   - No calibration targets (no April grids)
+   - Natural scene observations
+
+   Includes:
+   - Stereo images
+   - IMU data
+   - Ground-truth odometry
+
+   Purpose:
+   - Visual-inertial odometry (VIO)
+   - SLAM evaluation
+   - Sensor fusion benchmarking
+
+## Data Format
+
+Each example follows the same top-level schema. Some fields are split-dependent:
+
+- `stereo`: images + `timestamp` (IMU is stationary; pose not provided)
+- `stereoinertial`: adds IMU (`gyro`, `accel`) and time offsets (`sync_dt`)
+- `vio`: adds ground-truth pose (`position`, `orientation`)
+
+Example record:
+```json
+{
+  "image_left": { "path": "left/0.000000.png", "bytes": null },
+  "image_right": { "path": "right/0.000000.png", "bytes": null },
+  "timestamp": 0.0,
+
+  "gyro": [0.0, 0.0, 0.0],
+  "accel": [0.0, 0.0, 0.0],
+
+  "sync_dt": [0.0, 0.0],
+
+  "position": [0.0, 0.0, 0.0],
+  "orientation": [0.0, 0.0, 0.0, 1.0]
+}
+```
+
+Notes:
+- `gyro` is in rad/s and `accel` is in m/s^2.
+- `sync_dt = [dt_right, dt_imu]` are time offsets (in seconds) relative to `timestamp` (left image):
+  - `dt_right = t_right - t_left`
+  - `dt_imu = t_imu - t_left`
+
+### Sampling Methodology
+
+Each split contains 100 randomly sampled, synchronized frames:
+- Uniform sampling across the trajectory
+- Start/end trimmed to remove initialization artifacts
+
+Synchronization constraints:
+
+| Constraint | Threshold |
+| --- | --- |
+| `|t_left - t_right|` | `<= 5 ms` |
+| `|t_left - t_imu|` | `<= 5 ms` |
+| `|t_left - t_odom|` (VIO only) | `<= 5 ms` |
+
+### Missing Data Handling
+
+For splits without ground truth (stereo, stereoinertial):
+
+```text
+position    = [inf, inf, inf]
+orientation = [inf, inf, inf, inf]
+```
+
+### ROS Topics
+
+#### ROS 1 (Calibration)
+```text
+/stereo/left/color/image_raw
+/stereo/right/color/image_raw
+/imu/data
+```
+
+#### ROS 2 (VIO)
+```text
+/stereo/left/color/image_raw
+/stereo/right/color/image_raw
+/imu/data
+/ground_truth/odom
+/tf
+```
+
+## Example Usage
+
+```python
+from datasets import load_dataset
+import numpy as np
+
+ds = load_dataset("OmniInstrument/SLAM_project", split="vio")
+sample = ds[0]
+
+img_l = sample["image_left"]
+img_r = sample["image_right"]
+
+gyro = sample["gyro"]
+accel = sample["accel"]
+
+pos = sample["position"]
+quat = sample["orientation"]
+
+if not np.isinf(np.asarray(pos)).any():
+    print("Ground truth available")
+```
+
+## License
+This software and dataset are released under the [MIT License](LICENSE).

scripts/upload_preview_to_hf.py

@@ -0,0 +1,144 @@
+#!/usr/bin/env python3
+
+import argparse
+import math
+import os
+
+import pandas as pd
+from datasets import Dataset, Features, Image, Sequence, Value
+
+
+def parse_args():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--data", required=True, help="Path to preview folder")
+    parser.add_argument("--repo", required=True, help="HF dataset repo (e.g. org/name)")
+    parser.add_argument("--split", default="preview", help="Dataset split name")
+    return parser.parse_args()
+
+
+def _coerce_dt(x: object) -> float:
+    """Coerce merge-produced values (NaN/None/str) into a sane float offset."""
+    try:
+        v = float(x)
+    except (TypeError, ValueError):
+        return 0.0
+    if math.isnan(v) or math.isinf(v):
+        return 0.0
+    return v
+
+
+def main():
+    args = parse_args()
+
+    data_dir = args.data
+    imu_path = os.path.join(data_dir, "imu.csv")
+    odom_path = os.path.join(data_dir, "odom.csv")
+
+    if not os.path.exists(imu_path):
+        raise FileNotFoundError(f"imu.csv not found in {data_dir}")
+
+    print("Loading IMU data...")
+    imu = pd.read_csv(imu_path)
+
+    # Normalize timestamps (filename-friendly and merge-stable).
+    imu["t"] = imu["t"].map(lambda x: f"{float(x):.6f}")
+
+    has_odom = os.path.exists(odom_path)
+    if has_odom:
+        print("Loading ODOM data...")
+        odom = pd.read_csv(odom_path)
+        odom["t"] = odom["t"].map(lambda x: f"{float(x):.6f}")
+        df = pd.merge(imu, odom, on="t", how="left")
+    else:
+        print("No odom.csv found -> filling pose with inf")
+        df = imu.copy()
+
+    print(f"Rows after merge: {len(df)}")
+
+    left_paths = []
+    right_paths = []
+    timestamps = []
+    gyro = []
+    accel = []
+    sync_dt = []
+    position = []
+    orientation = []
+
+    missing = 0
+    for _, row in df.iterrows():
+        t = row["t"]
+
+        l = os.path.join(data_dir, "left", f"{t}.png")
+        r = os.path.join(data_dir, "right", f"{t}.png")
+
+        if not os.path.exists(l) or not os.path.exists(r):
+            missing += 1
+            continue
+
+        left_paths.append(l)
+        right_paths.append(r)
+        timestamps.append(float(t))
+
+        # IMU
+        gyro.append([row["gx"], row["gy"], row["gz"]])
+        accel.append([row["ax"], row["ay"], row["az"]])
+
+        # Sync offsets (seconds) relative to the left-image timestamp.
+        dt_r = _coerce_dt(row.get("dt_right", 0.0))
+        dt_i = _coerce_dt(row.get("dt_imu", 0.0))
+        sync_dt.append([dt_r, dt_i])
+
+        # ODOM (optional)
+        if has_odom:
+            position.append([row["px"], row["py"], row["pz"]])
+            orientation.append([row["qx"], row["qy"], row["qz"], row["qw"]])
+        else:
+            position.append([math.inf, math.inf, math.inf])
+            orientation.append([math.inf, math.inf, math.inf, math.inf])
+
+    if not left_paths:
+        raise RuntimeError("No valid samples found")
+
+    if missing:
+        print(f"Skipped {missing} rows due to missing images")
+
+    print(f"Final dataset size: {len(left_paths)}")
+
+    features = Features(
+        {
+            "image_left": Image(),
+            "image_right": Image(),
+            "timestamp": Value("float64"),
+            "gyro": Sequence(Value("float32"), length=3),
+            "accel": Sequence(Value("float32"), length=3),
+            "sync_dt": Sequence(Value("float32"), length=2),
+            "position": Sequence(Value("float32"), length=3),
+            "orientation": Sequence(Value("float32"), length=4),
+        }
+    )
+
+    data = {
+        "image_left": left_paths,
+        "image_right": right_paths,
+        "timestamp": timestamps,
+        "gyro": gyro,
+        "accel": accel,
+        "sync_dt": sync_dt,
+        "position": position,
+        "orientation": orientation,
+    }
+
+    print("Creating Hugging Face dataset...")
+    ds = Dataset.from_dict(data, features=features)
+
+    print(ds)
+
+    print(f"Pushing to {args.repo} (split={args.split})...")
+    ds.push_to_hub(args.repo, split=args.split)
+
+    print("Done.")
+
+
+if __name__ == "__main__":
+    main()
+