MVSFormer: Multi-View Stereo by Learning Robust Image Features and Temperature-based Depth

Open paper
Benchmark Model Rank Results
3d-reconstruction-on-dtuMVSFormer#3Overall: 0.289Acc: 0.327Comp: 0.251
point-clouds-on-tanks-and-templesMVSFormer#3Mean F1 (Advanced): 40.87Mean F1 (Intermediate): 66.37

Rank counts only results with a code link.