Add GlobalDiff inference data card
Browse files
README.md
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
pretty_name: GlobalDiff Inference Data
|
| 4 |
+
language:
|
| 5 |
+
- en
|
| 6 |
+
tags:
|
| 7 |
+
- co-speech-gesture-generation
|
| 8 |
+
- motion-generation
|
| 9 |
+
- diffusion-models
|
| 10 |
+
- beat2
|
| 11 |
+
- smpl-x
|
| 12 |
+
size_categories:
|
| 13 |
+
- n<1K
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
# GlobalDiff Inference Data
|
| 17 |
+
|
| 18 |
+
Generated SMPL-X predictions from the official GlobalDiff Speaker 2 release for comparison and visualization without rerunning the full generation pipeline.
|
| 19 |
+
|
| 20 |
+
- [Paper](https://arxiv.org/abs/2511.10076) · [Hugging Face Paper page](https://huggingface.co/papers/2511.10076)
|
| 21 |
+
- [Project page](https://xiangyuezhang.com/GlobalDiff/) · [Code](https://github.com/Xiangyue-Zhang/GlobalDiff)
|
| 22 |
+
- [Published version](https://doi.org/10.1609/aaai.v40i15.38281)
|
| 23 |
+
- [Model checkpoints](https://huggingface.co/X-Zhang/GlobalDiff)
|
| 24 |
+
|
| 25 |
+
## File
|
| 26 |
+
|
| 27 |
+
`best_pid_2.zip` contains the released generated test results for BEAT2 person 2.
|
| 28 |
+
|
| 29 |
+
## Provenance and terms
|
| 30 |
+
|
| 31 |
+
The archive contains generated predictions, not BEAT2 ground-truth motion or audio. Users remain responsible for complying with the licenses and terms of [BEAT2](https://huggingface.co/datasets/H-Liu1997/BEAT2), [SMPL-X](https://smpl-x.is.tue.mpg.de/), and the GlobalDiff repository. This card grants no additional rights.
|