search

shunsukesaito / PIFu

This repository contains the code for the paper "PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitization"
https://shunsukesaito.github.io/PIFu/
Other
1.76k stars 341 forks source link

@shunsukesaito SMPL models for the obj files generated #25

Closed PRAKHAR-bit closed 4 years ago

PRAKHAR-bit commented 4 years ago

@shunsukesaito Hi, thanks for the wonderful work. Can you tell me a way that the obj file which I generated using your code, I can convert it to a SMPL model. I can create an SMPL model with the arbitary parameters but I cannot texture it like PIFu . So any way in which I can use PIFu to generate a textured SMPL model

shunsukesaito commented 4 years ago

There is no straightforward way to convert PIFu reconstructions to SMPL as PIFu uses a non-parametric volumetric representation. One hack would be to reconstruct PIFu, fit SMPL model using HMR or whatever off-the-shelf reconstruction methods and then use only the color inference module in PIFu to texture SMPL. However, as the surface of PIFu and SMPL are not perfectly aligned, texture reconstruction on SMPL might be dirty.

PRAKHAR-bit commented 4 years ago

@shunsukesaito Thanks for replying .Special thanks for suggesting me alternate ways. For now decided to use dense pose to generate texture maps for SMPL textures Densepose makes them easily from 2d images. Waiting for PIFUhd. Regards.