|
Download foundationpose/README.md from erwtwert/embodied-data-loop-models: direct link, hf CLI and curl.
- Browser
- Download file 872 Bytes
-
https://huggingface.co/erwtwert/embodied-data-loop-models/resolve/main/foundationpose/README.md
- Command line
-
hf download hf://erwtwert/embodied-data-loop-models/foundationpose/README.md
-
curl -L -o README.md https://huggingface.co/erwtwert/embodied-data-loop-models/resolve/main/foundationpose/README.md
872 Bytes
| # FoundationPose refiner and scorer | |
| Upstream: https://github.com/NVlabs/FoundationPose . Original weight download: https://drive.google.com/drive/folders/1DFezOAD0oD1BblsXVxqDsl8fj0qzB82i . Copyright NVIDIA Corporation and affiliates. | |
| Both `2023-10-28-18-33-37` (refiner) and `2024-01-11-20-02-45` (scorer) contain `model_best.pth` and `config.yml`. These files use the [NVIDIA Source Code License](../licenses/FoundationPose.txt), limited to non-commercial research/evaluation; they are not covered by the companion project's MIT license. | |
| Companion GitHub script: [foundation_cloth.py](https://github.com/pm1255/human-robot-transfer/blob/main/human2robot/object_annotation/foundation_cloth.py). [Setup and usage](https://github.com/pm1255/human-robot-transfer/blob/main/docs/MODEL_RELEASE.md). Cloth output is a rigid global proxy, not deformable-object ground truth. | |