search

xiezhy6 / GP-VTON

Official Implementation for CVPR2023 paper "GP-VTON: Towards General Purpose Virtual Try-on via Collaborative Local-Flow Global-Parsing Learning"
475 stars 82 forks source link

Which human parser is most applicable for this task? #29

Open arunsubk opened 1 year ago

arunsubk commented 1 year ago

HI @xiezhy6 ,

Thanks a lot for the great work and sharing the same. Could you please let me know which human parser is capable of giving the cloth part of the arm and non-cloth part of the arm? The best I know the existing parsers, either provide the top part of the arm and the bottom part of an arm, irrespective of whether the cloth is there or not (for example Pascal), or provide arm versus trunk. While I can imagine thinking of an intersection, it looks too tedious. Could you please let me know if your human parser is also developed on your own? Else, could you please help with pointing to which human parser was used?

Thanks,

tobias861222 commented 1 year ago

Has anyone found a human parser that can be used with GPVTON ?