DisCo
DisCo copied to clipboard
[CVPR2024] DisCo: Referring Human Dance Generation in Real World
多机多卡的运行脚本,谢谢
 Above is the dataset used by you on GitHub. However, I found that you also used the MoreTikTok dataset in the paper:  Are these two datasets the same?
Hi, thanks for the great work. I want to train with the `512*512` dataset, and I set `--image_size` to `512`, but it doesn't work.
Hi, thanks for the great work. I noticed that the LPIPS evaluation does not include `normalize=True` given that the inputs are in the [0,1] range. Adding this would change the...
I had some problems with jax and jaxlib latest version for running colab code so I used pip install jax==0.4.23 pip install jaxlib==0.4.23 which are compatible with each other.
new10val_TiktokDance-poses-masks.caption.linelist.lineidx new10val_TiktokDance-poses-masks.caption.linelist.lineidx.8b new10val_TiktokDance-poses-masks.caption.linelist.tsv new10val_captions.lineidx new10val_captions.lineidx.8b new10val_captions.lineidx.tsv
How can I get "More TikTok-Style Training Data" please?
I find that in your new arxiv paper, you add the temporal modeling and further improve the quantitative metrics. Do you have a plan to release the new model weight...