vietanhdev
/

segment-anything-2.1-onnx-models

@@ -1,106 +1,92 @@
----
-license: apache-2.0
-tags:
-  - image-segmentation
-  - segment-anything
-  - segment-anything-2
-  - onnx
-  - onnxruntime
-library_name: onnxruntime
----
-# Segment Anything 2.1 (SAM 2.1) — ONNX Models
-ONNX-exported versions of Meta's **Segment Anything Model 2.1 (SAM 2.1)** — an improved version of [SAM 2](https://github.com/facebookresearch/segment-anything-2) with better accuracy and robustness — ready for CPU/GPU inference with [ONNX Runtime](https://onnxruntime.ai/).
-These models are used by **[AnyLabeling](https://github.com/vietanhdev/anylabeling)** for AI-assisted image annotation, and exported by **[samexporter](https://github.com/vietanhdev/samexporter)**.
-## Available Models
-| File | Variant | Notes |
-|------|---------|-------|
-| `sam2.1_hiera_tiny_20260221.zip` | SAM 2.1 Hiera-Tiny | Smallest, fastest |
-| `sam2.1_hiera_small_20260221.zip` | SAM 2.1 Hiera-Small | Good balance |
-| `sam2.1_hiera_base_plus_20260221.zip` | SAM 2.1 Hiera-Base+ | Higher accuracy |
-| `sam2.1_hiera_large_20260221.zip` | SAM 2.1 Hiera-Large | Most accurate |
-Each zip contains two ONNX files: an **encoder** (runs once per image) and a **decoder** (runs interactively for each prompt).
-## What's Improved vs SAM 2?
-SAM 2.1 offers improved segmentation accuracy and better handling of edge cases compared to SAM 2. The ONNX conversion process is identical, and both use the same inference API.
-## Prompt Types
-- **Point** (`+point` / `-point`): click to include/exclude regions
-- **Rectangle**: draw a bounding box around the target object
-## Use with AnyLabeling (Recommended)
-[AnyLabeling](https://github.com/vietanhdev/anylabeling) is a desktop annotation tool with a built-in model manager that downloads, caches, and runs these models automatically — no coding required.
-1. Install: `pip install anylabeling`
-2. Launch: `anylabeling`
-3. Click the **Brain** button → select a **Segment Anything 2.1** model from the dropdown
-4. Use point or rectangle prompts to segment objects
-[![AnyLabeling demo](https://user-images.githubusercontent.com/18329471/236625792-07f01838-3f69-48b0-a12e-30bad27bd921.gif)](https://github.com/vietanhdev/anylabeling)
-## Use Programmatically with ONNX Runtime
-```python
-import urllib.request, zipfile
-url = "https://huggingface.co/vietanhdev/segment-anything-2.1-onnx-models/resolve/main/sam2.1_hiera_tiny_20260221.zip"
-urllib.request.urlretrieve(url, "sam2.1_hiera_tiny.zip")
-with zipfile.ZipFile("sam2.1_hiera_tiny.zip") as z:
-    z.extractall("sam2.1_hiera_tiny")
-```
-Then use [samexporter](https://github.com/vietanhdev/samexporter)'s inference module:
-```bash
-pip install samexporter
-python -m samexporter.inference \
-    --encoder_model sam2.1_hiera_tiny/sam2.1_hiera_tiny.encoder.onnx \
-    --decoder_model sam2.1_hiera_tiny/sam2.1_hiera_tiny.decoder.onnx \
-    --image photo.jpg \
-    --prompt prompt.json \
-    --output result.png \
-    --sam_variant sam2
-```
-## Re-export from Source
-To re-export or customize the models using [samexporter](https://github.com/vietanhdev/samexporter):
-```bash
-pip install samexporter
-pip install git+https://github.com/facebookresearch/segment-anything-2.git
-# Download SAM 2.1 checkpoints
-bash download_all_models.sh
-# Export Tiny variant
-python -m samexporter.export_sam2 \
-    --checkpoint original_models/sam2.1_hiera_tiny.pt \
-    --output_encoder output_models/sam2.1_hiera_tiny.encoder.onnx \
-    --output_decoder output_models/sam2.1_hiera_tiny.decoder.onnx \
-    --model_type sam2.1_hiera_tiny
-# Or convert all SAM 2 and SAM 2.1 variants at once:
-bash convert_all_meta_sam2.sh
-```
-## Related Repositories
-| Repo | Description |
-|------|-------------|
-| [vietanhdev/samexporter](https://github.com/vietanhdev/samexporter) | Export scripts, inference code, conversion tools |
-| [vietanhdev/anylabeling](https://github.com/vietanhdev/anylabeling) | Desktop annotation app powered by these models |
-| [vietanhdev/segment-anything-2-onnx-models](https://huggingface.co/vietanhdev/segment-anything-2-onnx-models) | Original SAM 2 ONNX models |
-| [facebookresearch/segment-anything-2](https://github.com/facebookresearch/segment-anything-2) | Original SAM 2 / SAM 2.1 by Meta |
-## License
-The ONNX models are derived from Meta's SAM 2.1, released under the **Apache 2.0** license.
-The export code is part of [samexporter](https://github.com/vietanhdev/samexporter), released under the **MIT** license.

+---
+license: apache-2.0
+pipeline_tag: image-segmentation
+library_name: onnx
+tags:
+  - onnxruntime
+  - onnx
+  - segment-anything
+  - segment-anything-2
+  - segment-anything-2.1
+  - image-segmentation
+  - edge-ai
+  - anylabeling
+authors:
+  - Viet-Anh Nguyen
+---
+# Segment Anything 2.1 (SAM 2.1) — ONNX Models
+ONNX exports of Meta's [SAM 2.1](https://github.com/facebookresearch/sam2) backbones — the maintenance release of SAM 2 — packaged for direct use with [`onnxruntime`](https://onnxruntime.ai) and [AnyLabeling](https://github.com/vietanhdev/anylabeling).
+## Why this repo exists
+SAM 2.1 ships incremental quality improvements over SAM 2 with the same architecture and runtime cost. ONNX gives you a portable, dependency-light runtime that works in Python, C++, JavaScript, and most embedded targets. These exports are the ones AnyLabeling consumes for its smart-labeling features.
+## Variants
+Each `.zip` bundles the encoder + decoder ONNX files for that backbone. The `_20260221` suffix is the export date.
+| File | Backbone | Size |
+|---|---|---|
+| `sam2.1_hiera_tiny_20260221.zip` | Hiera-T | 111 MB |
+| `sam2.1_hiera_small_20260221.zip` | Hiera-S | 136 MB |
+| `sam2.1_hiera_base_plus_20260221.zip` | Hiera-B+ | 259 MB |
+| `sam2.1_hiera_large_20260221.zip` | Hiera-L | 768 MB |
+## Quick start
+```bash
+pip install huggingface_hub onnxruntime
+```
+```python
+from huggingface_hub import hf_hub_download
+import zipfile, onnxruntime as ort
+zip_path = hf_hub_download(repo_id="vietanhdev/segment-anything-2.1-onnx-models",
+                           filename="sam2.1_hiera_tiny_20260221.zip")
+with zipfile.ZipFile(zip_path) as z:
+    z.extractall("./sam21_tiny")
+enc = ort.InferenceSession("./sam21_tiny/encoder.onnx", providers=["CPUExecutionProvider"])
+print([(i.name, i.shape, i.type) for i in enc.get_inputs()])
+```
+For the full image → mask pipeline (encoder + decoder + prompt handling), see how AnyLabeling wires it: <https://github.com/vietanhdev/anylabeling>
+## Use with AnyLabeling
+These models drop into AnyLabeling's auto-labeling backend without conversion. See the [AnyLabeling docs](https://github.com/vietanhdev/anylabeling) for the model-config wiring.
+## Source weights
+Original SAM 2.1 weights and license: <https://github.com/facebookresearch/sam2>
+This repo redistributes the same weights in ONNX format. License unchanged from upstream (Apache 2.0).
+## Citation
+```bibtex
+@misc{nguyen2026sam21_onnx,
+  author = {Nguyen, Viet-Anh and {Neural Research Lab}},
+  title  = {SAM 2.1 ONNX Models},
+  year   = {2026},
+  url    = {https://huggingface.co/vietanhdev/segment-anything-2.1-onnx-models}
+}
+```
+For the underlying model, cite Meta's SAM 2 paper (SAM 2.1 is a release of the same model family):
+```bibtex
+@article{ravi2024sam2,
+  title   = {SAM 2: Segment Anything in Images and Videos},
+  author  = {Ravi, Nikhila and others},
+  journal = {arXiv:2408.00714},
+  year    = {2024}
+}
+```
+## Acknowledgments
+Thanks to Meta AI Research for the SAM 2.x line. This repo packages their work for edge inference.