search

fyviezhao / M3D-VTON

Official code for ICCV2021 paper "M3D-VTON: A Monocular-to-3D Virtual Try-on Network"
166 stars 38 forks source link

mesh results #19

Open sanazsab opened 2 years ago

sanazsab commented 2 years ago

Hi,

Thanks for your work.

When I used your images with the same resolution, the output of rgbd2pcl which is .ply is not completely painted and there are some parts that are masked. Is there anything that should apply after the rgbd2pck before mesh lab?

fyviezhao commented 2 years ago

I believe the masked part is due to the mis-parsed region of the original image such as bags. You can use this parsing method for higher parsing granularity which further supports bags, belt, etc.

The output point cloud from rgbd2pcd.py can be directly sent to mesh lab for remeshing without other processing.

BhaumikThakkar commented 2 years ago

I believe the masked part is due to the mis-parsed region of the original image such as bags. You can use this parsing method for higher parsing granularity which further supports bags, belt, etc.

The output point cloud from rgbd2pcd.py can be directly sent to mesh lab for remeshing without other processing.

Hello there. Is it okay to use this Parsing method(self correction) instead of 2d human parsing?