Preprocessing code for a custom video dataset
Oocean35 opened this issue · comments
Oocean35 commented
Hi @skhu101 ,
Thanks a lot for sharing this very insightful work !
It would be really great if you could provide the code that one can use to prepare one's own video dataset to use for training !
This would be incredibly helpful !
Thanks a lot !
Shoukang Hu commented
Hi, I think for the video dataset, you need to first estimate the camera and SMPL parameters. Maybe one possible solution is to use CLIFF (CLIFF: Carrying Location Information in Full Frames into Human Pose and Shape Estimation). Then you can follow our RenderPeople dataloader to train the model.