JeremyCJM / DiffSHEG

[CVPR'24] DiffSHEG: A Diffusion-Based Approach for Real-Time Speech-driven Holistic 3D Expression and Gesture Generation
https://jeremycjm.github.io/proj/DiffSHEG/
BSD 3-Clause "New" or "Revised" License
112 stars 9 forks source link

Test on SHOW dataset #17

Open Mumuwei opened 2 months ago

Mumuwei commented 2 months ago

Hello, when I ran Test on SHOW dataset, I encountered the following error, I did not find the corresponding code, how to correct this problem? image

JeremyCJM commented 2 months ago

The code here is using the original evaluation code of BEAT to evaluate the generated motion. You can first try running the code of BEAT and then adapting to SHOW data.