Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
tags:
|
| 4 |
+
- diffusion
|
| 5 |
+
- reinforcement-learning
|
| 6 |
+
- human-motion-generation
|
| 7 |
+
- physics-based-character-control
|
| 8 |
+
- isaac-gym
|
| 9 |
+
---
|
| 10 |
+
Official model checkpoints for **NaP-Control: Navigating Diffusion Prior for Versatile and Fast Character Control**
|
| 11 |
+
|
| 12 |
+
[](https://github.com/chiawenchen/NaP)
|
| 13 |
+
[](https://arxiv.org/abs/2605.20209)
|
| 14 |
+
[](https://chiawenchen.github.io/nap-control-project/)
|
| 15 |
+
|
| 16 |
+
---
|
| 17 |
+
## 🚀 Quick Start & Download
|
| 18 |
+
|
| 19 |
+
### Option 1: Automated Download (Recommended)
|
| 20 |
+
The easiest way to get started is to use the setup script provided in the main GitHub repository. Follow the README.md and run:
|
| 21 |
+
```bash
|
| 22 |
+
bash download_data.sh
|
| 23 |
+
```
|
| 24 |
+
This script will automatically fetch all checkpoints from Hugging Face and place them into their correct directory paths.
|
| 25 |
+
## Option 2: Manual Download & Organization
|
| 26 |
+
If you prefer to download the files manually, you must arrange them into the following directory structure inside your workspace root:
|
| 27 |
+
```
|
| 28 |
+
|-- assets
|
| 29 |
+
|-- data
|
| 30 |
+
|-- nap
|
| 31 |
+
|-- output
|
| 32 |
+
|-- HumanoidIm
|
| 33 |
+
|-- agile_goal
|
| 34 |
+
|-- Humanoid.pth
|
| 35 |
+
|-- agile_goal_terrain
|
| 36 |
+
|-- Humanoid.pth
|
| 37 |
+
|-- far_goal
|
| 38 |
+
|-- Humanoid.pth
|
| 39 |
+
|-- far_goal_terrain
|
| 40 |
+
|-- Humanoid.pth
|
| 41 |
+
|-- multi_goal
|
| 42 |
+
|-- Humanoid.pth
|
| 43 |
+
|-- pulse_vae_iclr
|
| 44 |
+
|-- sit
|
| 45 |
+
|-- Humanoid.pth
|
| 46 |
+
|-- traj
|
| 47 |
+
|-- Humanoid.pth
|
| 48 |
+
|-- velocity
|
| 49 |
+
|-- Humanoid.pth
|
| 50 |
+
|-- velocity_terrain
|
| 51 |
+
|-- Humanoid.pth
|
| 52 |
+
...
|
| 53 |
+
|-- UniPhys
|
| 54 |
+
...
|
| 55 |
+
|-- isaac_utils
|
| 56 |
+
|-- output
|
| 57 |
+
|-- HumanoidIm
|
| 58 |
+
|-- root_with_dof
|
| 59 |
+
|-- checkpoints
|
| 60 |
+
|-- last.ckpt
|
| 61 |
+
|-- train_data_stats.npy
|
| 62 |
+
...
|
| 63 |
+
```
|
| 64 |
+
## 📜 Citation
|
| 65 |
+
If you find our work, code, or checkpoints useful for your research, please cite our paper:
|
| 66 |
+
```bibtex
|
| 67 |
+
@misc{chen2026napcontrolnavigatingdiffusionprior,
|
| 68 |
+
title={NaP-Control: Navigating Diffusion Prior for Versatile and Fast Character Control},
|
| 69 |
+
author={Chia-Wen Chen and Yan Wu and Korrawe Karunratanakul and Siyu Tang},
|
| 70 |
+
year={2026},
|
| 71 |
+
eprint={2605.20209},
|
| 72 |
+
archivePrefix={arXiv},
|
| 73 |
+
primaryClass={cs.GR},
|
| 74 |
+
url={https://arxiv.org/abs/2605.20209},
|
| 75 |
+
}
|
| 76 |
+
```
|