MiDaS 3.1

Reported on 4 benchmarks across 2 tasks · 1 paper · 2 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision2 results

Depth EstimationonETH3D
Delta < 1.25· 2023-07-26
0.867
best: 0.981 (Distill Any Depth)
SOTA
MiDaS v3.1 -- A Model Zoo for Robust Monocular Relative Depth Estimation arXiv:2307.14460
Depth EstimationonETH3D
absolute relative error· 2023-07-26
0.0139
best: 0.0121 (HDN)
MiDaS v3.1 -- A Model Zoo for Robust Monocular Relative Depth Estimation arXiv:2307.14460

Methodology2 results

3DonETH3D
Delta < 1.25· 2023-07-26
0.867
best: 0.981 (Distill Any Depth)
SOTA
MiDaS v3.1 -- A Model Zoo for Robust Monocular Relative Depth Estimation arXiv:2307.14460
3DonETH3D
absolute relative error· 2023-07-26
0.0139
best: 0.0121 (HDN)
MiDaS v3.1 -- A Model Zoo for Robust Monocular Relative Depth Estimation arXiv:2307.14460