fraunhoferhhi / neural-deferred-shading

Multi-View Mesh Reconstruction with Neural Deferred Shading (CVPR 2022)

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Details about optimization

boqian-li opened this issue · comments

Hi, i wanna know during the training and optimization process, are there the positions of the vertices adjusted? or each time it just fit a new mesh with new vertices and triangles?

Hi, each scene is individually reconstructed, which means the method always starts from some template mesh and then moves the vertex positions during optimization.