Pretrained model for "Toward Practical Monocular Indoor Depth Estimation" (DistDepth, CVPR 2022)
Latest
Pretrained models for "Toward Practical Monocular Indoor Depth Estimation" (DistDepth, CVPR 2022) by Cho-Ying Wu, Jialiang Wang, Michael Hall, Ulrich Neumann, Shuochen Su.
- Official implementation and pre-trained models: https://github.com/facebookresearch/DistDepth
- Small changes to export via torch script tracing.
- The exported model currently fails with GPU inference, so only CPU inference is enabled