arxiv.org
scholar.google.com
Everybody Dance Now
Caroline Chan and Shiry Ginosar and Tinghui Zhou and Alexei A. Efros
arXiv e-Print archive - 2018 via Local arXiv
Keywords: cs.GR, cs.CV

more

[link]
Summary by Oleksandr Bailo 2 years ago
Loading...
Have people attempted to reproduce? Is there a piece of code somewhere with the end-to-end, or do I need to put the different pieces together (i.e. OpenPose, pix2pixHD, FaceGAN)?

Yes, I have reproduced the code (combined different pieces together), and it works very well. It is basically [pix2pixHD](https://github.com/NVIDIA/pix2pixHD) with temporal smoothing. For FaceGAN, you could just use a global generator from pix2pixHD model (instead of local and global generators together). To get the poses, I have utilized out-of-box executables provided by [OpenPose](https://github.com/CMU-Perceptual-Computing-Lab/openpose).

Your comment:


ShortScience.org allows researchers to publish paper summaries that are voted on and ranked!
About

Sponsored by: