wedetect-tiny-onnx / README.md
sweetlhare's picture
Upload README.md with huggingface_hub
c7cc9bd verified
|
Raw
History Blame Contribute Delete
2.15 kB
---
license: gpl-3.0
library_name: onnxruntime
pipeline_tag: object-detection
tags:
- open-vocabulary-detection
- onnx
- webgpu
- browser
- wedetect
---
# WeDetect-tiny β€” browser ONNX (int8)
Browser-ready **int8 ONNX** export of **WeDetect-tiny**, an open-vocabulary object detector
(text-prompted boxes). Two towers run client-side via `onnxruntime-web` (WebGPU, WASM fallback):
| file | size | role |
|---|---|---|
| `wedetect_tiny_language_int8.onnx` | ~279 MB | XLM-R text encoder β†’ prompt features (once per prompt set) |
| `wedetect_tiny_vision_int8.onnx` | ~38 MB | image encoder + detection head (per image) |
Pipeline: prompts β†’ XLM-R tokenizer (`Xenova/xlm-roberta-base`) β†’ `language.onnx` β†’ text features;
image β†’ letterbox 640 β†’ `vision.onnx` β†’ decoded xyxy + sigmoid scores β†’ score filter + class-agnostic
NMS + rescale.
## License β€” GPL-3.0
WeDetect is **GPL-3.0**. These ONNX weights are a derived export of the upstream GPL model and are
therefore distributed under **GPL-3.0-only**. Using them keeps you within GPL terms.
**Upstream / attribution:** WeDetect by **WeChatCV** β€” https://github.com/WeChatCV/WeDetect
(GPL-3.0). This repository redistributes an ONNX/int8 export of that model with no change to its
license. Corresponding source (model definition + training/export pipeline) lives in the upstream
project; the browser export/decoder is in the momoto-plugins repo (also GPL-3.0):
https://github.com/sweetlhare/momoto-plugins (`src/copyleft/wedetect.js`).
## How it's used
Loaded at runtime by the **momoto** WeDetect plugin (`momoto-plugins/src/copyleft/wedetect.js`),
which points `meta.assets` at the two ONNX URLs here. The plugin is opt-in and never bundled into a
closed build β€” GPL bytes are fetched from this public GPL host at the user's choice, keeping the
closed core clear of copyleft. Self-hosters can override the asset URLs.
Resolve URLs (once uploaded):
- `https://huggingface.co/sweetlhare/wedetect-tiny-onnx/resolve/main/wedetect_tiny_vision_int8.onnx`
- `https://huggingface.co/sweetlhare/wedetect-tiny-onnx/resolve/main/wedetect_tiny_language_int8.onnx`