BDMM icon indicating copy to clipboard operation
BDMM copied to clipboard

Official PyTorch implementation of BDMM: Bidirectionally Deformable Motion Modulation For Video-based Human Pose Transfer [ICCV 2023]

Results 4 BDMM issues
Sort by recently updated
recently updated
newest added

I run train.py with your code but I ve got Nan in loss graph. I did not change dataset and any setting. ![image](https://github.com/rocketappslab/BDMM/assets/67573223/e76b047f-a4dc-4755-8355-527de7e49bc6)

when i run train.py ,i have a error: ![image](https://github.com/rocketappslab/BDMM/assets/48466610/4ae5a24e-b009-4b28-8378-26f314e07675) i print self.pre_warp_feat_iter ,i find this ![image](https://github.com/rocketappslab/BDMM/assets/48466610/34c5b5ee-56dc-4fd7-b2ea-205482ac8a2d) i find define : nxt_warp_feat_iter = [None, feat_warp64_state, feat_warp128_state],but None have no detach ![image](https://github.com/rocketappslab/BDMM/assets/48466610/4c0f002e-a4fd-4b22-acbd-d481d59941d0)

This repository uses the following code in this [file](https://github.com/rocketappslab/BDMM/blob/main/evaluation/getMetrics_animation.py). But the formula should be `10 * math.log10(255.0 / math.sqrt(mse))`, The result of the calculation is 2 times larger than the...

Thanks for your great work. I want to know wheather getMetrics_animation.py can eval the performance on other datasets such as Tiktok dataset. Hope for your reply. Thank you!