Web page for the Fashion dataset introduced in the BMVC2019 paper “DwNet: Dense warp-based network for pose-guided human video generation”.
Here is our demo video for both Fashion dataset and Taichi:
Here some samples from the Fashion dataset:
To obtain the train data and the test data you need to run the script from the same directory where you put the above files.
© 2019 UBC Computer Vision Lab. Site adapted from here; copy and modify it for your own use.