MUST-GAN
MUST-GAN copied to clipboard
Pytorch implementation of CVPR2021 paper "MUST-GAN: Multi-level Statistics Transfer for Self-driven Person Image Generation"
  
I notice that the pose encoder are used Pa and Pa_con as input, I want to know the difference of only using Pa as input or Pa_con. Because Pa and...
i.e., the .png files and the .npy files in semantic_merge3
If the number of key points is not 18, that is, the number of channels is greater than 18, which part of the code needs to be changed?
Thanks for you great work, would you please provide you quantitive accseement code ,like IS,SSIM and FID?
I run the test.sh with 1 GPU, the output is a grey image with no person. But it works with 2 GPUs, is there anyway to run with 1 GPU...
Hello, I would like to ask for the decompression password. Can you please explain it