Upload folder using huggingface_hub

2023-10-28-18-33-37/config.yml

@@ -0,0 +1,39 @@
+lr: 0.0001
+c_in: 6
+zfar: 'Infinity'
+debug: null
+w_rot: 0.1
+n_view: 1
+run_id: null
+use_BN: true
+rot_rep: axis_angle
+ckpt_dir: null
+exp_name: 2023-10-28-18-33-37
+save_dir: /tmp/2023-10-28-18-33-37/
+loss_type: l2
+optimizer: adam
+trans_rep: tracknet
+batch_size: 64
+crop_ratio: 1.2
+use_normal: false
+BN_momentum: 0.1
+max_num_key: null
+warmup_step: -1
+input_resize:
+- 160
+- 160
+max_step_val: 1000
+normal_uint8: false
+vis_interval: 1000
+weight_decay: 0
+n_max_objects: null
+normalize_xyz: true
+clip_grad_norm: 'Infinity'
+rot_normalizer: 0.3490658503988659
+trans_normalizer:
+- 0.019999999552965164
+- 0.019999999552965164
+- 0.05000000074505806
+max_step_per_epoch: 25000
+val_epoch_interval: 10
+n_dataloader_workers: 60

2023-10-28-18-33-37/model_best.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:774700586ddc435d408fc01c9809c43e151232936369dfbea0f0f964ba471d60
+size 68220109

2024-01-11-20-02-45/config.yml

@@ -0,0 +1,41 @@
+lr: 0.0001
+c_in: 6
+zfar: 'Infinity'
+debug: null
+n_view: 1
+run_id: 3wy8qqex
+use_BN: true
+exp_name: 2024-01-11-20-02-45
+n_epochs: 62
+save_dir: /home/bowenw/debug/2024-01-11-20-02-45/
+use_mask: false
+loss_type: pairwise_valid
+optimizer: adam
+batch_size: 64
+crop_ratio: 1.1
+enable_amp: true
+use_normal: false
+max_num_key: null
+warmup_step: -1
+input_resize:
+- 160
+- 160
+max_step_val: 1000
+vis_interval: 1000
+weight_decay: 0
+normalize_xyz: true
+resume_run_id: null
+clip_grad_norm: 'Infinity'
+lr_epoch_decay: 500
+render_backend: nvdiffrast
+train_num_pair: 5
+lr_decay_epochs:
+- 50
+n_epochs_warmup: 1
+make_pair_online: false
+gradient_max_norm: 'Infinity'
+max_step_per_epoch: 10000
+n_rendering_workers: 1
+save_epoch_interval: 100
+n_dataloader_workers: 100
+split_objects_across_gpus: true

2024-01-11-20-02-45/model_best.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81924d384bf5c26c646ee4783104982ae3d1e049c181c36641b6a7aeae494c26
+size 190229389

README.md

@@ -0,0 +1,124 @@
+---
+license: cc-by-nc-4.0
+tags:
+  - computer-vision
+  - 6d-pose-estimation
+  - object-detection
+  - robotics
+  - foundationpose
+library_name: foundationpose
+---
+
+# FoundationPose Model Weights
+
+Pre-trained weights for [FoundationPose](https://github.com/NVlabs/FoundationPose) 6D object pose estimation model.
+
+## Model Details
+
+- **Refiner weights:** `2023-10-28-18-33-37/model_best.pth`
+- **Scorer weights:** `2024-01-11-20-02-45/model_best.pth`
+- **Source:** [Official FoundationPose release](https://github.com/NVlabs/FoundationPose)
+- **Paper:** [FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects (CVPR 2024)](https://arxiv.org/abs/2312.08344)
+
+## Model Architecture
+
+FoundationPose is a unified foundation model for 6D object pose estimation and tracking, supporting both:
+- **Model-based setup**: Using CAD models
+- **Model-free setup**: Using reference images (16-20 views)
+
+## Files
+
+```
+.
+├── 2023-10-28-18-33-37/
+│   ├── config.yml
+│   └── model_best.pth (refiner model)
+└── 2024-01-11-20-02-45/
+    ├── config.yml
+    └── model_best.pth (scorer model)
+```
+
+## Usage
+
+### Download Weights
+
+```python
+from huggingface_hub import snapshot_download
+
+# Download all weights
+weights_path = snapshot_download(
+    repo_id="gpue/foundationpose-weights",
+    local_dir="./weights"
+)
+```
+
+### Use with FoundationPose Space
+
+This model repository is designed to work with the [gpue/foundationpose](https://huggingface.co/spaces/gpue/foundationpose) Space.
+
+Set environment variables:
+```bash
+FOUNDATIONPOSE_MODEL_REPO=gpue/foundationpose-weights
+USE_HF_WEIGHTS=true
+USE_REAL_MODEL=true
+```
+
+### Local Usage
+
+```python
+import torch
+from pathlib import Path
+
+# Load refiner
+refiner_weights = torch.load("weights/2023-10-28-18-33-37/model_best.pth")
+
+# Load scorer
+scorer_weights = torch.load("weights/2024-01-11-20-02-45/model_best.pth")
+```
+
+## Performance
+
+- **Accuracy**: State-of-the-art on BOP benchmark (as of 2024/03)
+- **Speed**: Real-time capable with GPU acceleration
+- **Generalization**: Works on novel objects without fine-tuning
+
+## Citation
+
+If you use these weights, please cite:
+
+```bibtex
+@inproceedings{wen2023foundationpose,
+  title={FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects},
+  author={Wen, Bowen and Yang, Wei and Kautz, Jan and Birchfield, Stan},
+  booktitle={CVPR},
+  year={2024}
+}
+```
+
+## License
+
+These weights are from the official FoundationPose release and are subject to NVIDIA's [Source Code License](https://github.com/NVlabs/FoundationPose/blob/main/LICENSE.txt).
+
+**Key restrictions:**
+- Non-commercial use only
+- No redistribution of derivative works
+- Academic and research purposes
+
+## Related Resources
+
+- **Paper**: https://arxiv.org/abs/2312.08344
+- **Code**: https://github.com/NVlabs/FoundationPose
+- **Project Page**: https://nvlabs.github.io/FoundationPose/
+- **Inference Space**: https://huggingface.co/spaces/gpue/foundationpose
+
+## Model Card
+
+**Developed by:** NVIDIA Research (Bowen Wen, Wei Yang, Jan Kautz, Stan Birchfield)
+
+**Model type:** Transformer-based 6D pose estimator
+
+**Training data:** Large-scale synthetic dataset
+
+**Intended use:** 6D object pose estimation and tracking for robotics and AR/VR applications
+
+**Out-of-scope:** Commercial deployment (due to license restrictions)