| Author |
Message |
![]() ![]() ![]() ![]()
Jambalahaart_raja
Side Hero Username: Jambalahaart_raja
Post Number: 8976 Registered: 07-2008 Posted From: 69.115.176.215
Rating: N/A Votes: 0 (Vote!) | | Posted on Friday, August 24, 2018 - 12:53 am: |
![]() ![]() ![]() ![]() ![]() |
Biggboss:spatio-temporal smoothing
Biggboss:temporally coherent video generation including realistic face synthesis.
Pre-Prod Business ki PPT laa vundi!!! Long way to go for grace, naturality, movement completion etc etc.. possibly never... "Chill Bro. I told you to let it go!!" - The Budhha. |
![]() ![]() ![]() ![]()
Biggboss
Side Hero Username: Biggboss
Post Number: 4912 Registered: 08-2017 Posted From: 24.6.111.180
Rating: N/A Votes: 0 (Vote!) | | Posted on Friday, August 24, 2018 - 12:43 am: |
![]() ![]() ![]() ![]() ![]() |
https://twitter.com/hardmaru/status/1032762806796312576 https://arxiv.org/abs/1808.07371 Everybody Dance Now Caroline Chan, Shiry Ginosar, Tinghui Zhou, Alexei A. Efros (Submitted on 22 Aug 2018) This paper presents a simple method for "do as I do" motion transfer: given a source video of a person dancing we can transfer that performance to a novel (amateur) target after only a few minutes of the target subject performing standard moves. We pose this problem as a per-frame image-to-image translation with spatio-temporal smoothing. Using pose detections as an intermediate representation between source and target, we learn a mapping from pose images to a target subject's appearance. We adapt this setup for temporally coherent video generation including realistic face synthesis. |
|