The data set includes;
- 450 scenes with six camera angles each (see below)
- Ground-truth depth
- Instance + material segmentation
- Camera intrinsic / extrinsics
We experimented with ToteMVS on a customized version of MVSFormer, which can be found here.
The data set includes;
We experimented with ToteMVS on a customized version of MVSFormer, which can be found here.