|
Download README.md from EclipseAidge/SuperPoint: direct link, hf CLI and curl.
- Browser
- Download file 774 Bytes
-
https://huggingface.co/EclipseAidge/SuperPoint/resolve/main/README.md
- Command line
-
hf download hf://EclipseAidge/SuperPoint/README.md
-
curl -L -o README.md https://huggingface.co/EclipseAidge/SuperPoint/resolve/main/README.md
774 Bytes
| license: mit | |
| library_name: onnx | |
| tags: | |
| - onnx | |
| - aidge | |
| # SuperPoint ONNX | |
| SuperPoint model exported to ONNX by PrunaAI. The base SuperPoint model is adapted from https://github.com/rpautrat/SuperPoint repository `superpoint_pytorch.py` script and also uses the corresponding weights from the repository. | |
| ## Model | |
| - Source model: SuperPoint | |
| - Export format: ONNX | |
| - ONNX opset: 18 | |
| - Input: `inputs`, shape `[1, 1, 768, 1024]` | |
| - Output: `keypoints`, shape `[1, 512, 2]`; `scores`, shape `[1, 512]`; `descriptors`, shape `[1, 512, 256]` | |
| ## Usage in Aidge | |
| The model is intended to be used as the base model in the Aidge SuperPoint pruning tutorial. | |
| ## License | |
| The original SuperPoint model is provided in https://github.com/rpautrat/SuperPoint under MIT license. | |