ReMoDiffuse
ReMoDiffuse copied to clipboard
ReMoDiffuse: Retrieval-Augmented Motion Diffusion Model
1. You only put the visual code of human ml3d, I want to change the visual code of kit data set, the result of running out is particularly poor, and...
Could the visualization code output the 3D format? Now it could output gif format as example.
Hi Author, Thanks for your excellent work. How to get tokens in data folder? Thanks
Hi Dr.Zhang! Thanks for your great work! However, I can't find the training detail of FID minimizing in the Condition Mixture part. I'd really appreciated if you could help point...
Hi, thanks for your great work! But as in your README. It seems that I can only output the mp4 files or obj files. Can we output the skeleton files...
Thanks for your excellent work. Could you provide weights trained on the HumanML3D dataset?
Hi! I used your implementation of MDM with the following command: ``` python3 tools/dist_train.sh configs/mdm/mdm_t2m_official.py logs/mdm 8 --no-validate ``` The result was pretty bad with an FID of 41. However,...
Hi, thanks for your great work and code. I have one question. Will you release the code of the interface module, which uses different avatars to visualize the motion?
Where is the training condition mixture part? Is it not used for training? Please explain.
Hi, mingyuan, thanks for your excellent work. When I read the code, I find the motion-length in test stage is obtained via the `length = len(motion)` in `Pipeline Module: Crop`...