GMFlowNet

Reported on 6 benchmarks across 1 task · 1 paper

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

Computer Vision6 results

Optical Flow EstimationonSintel-clean
Average End-Point Error· 2022-03-21
1.39
best: 0.963 (MEMFOF-L)
Global Matching with Overlapping Attention for Optical Flow Estimation arXiv:2203.11335
Optical Flow EstimationonSintel-final
Average End-Point Error· 2022-03-21
2.648
best: 1.907 (MEMFOF-L)
Global Matching with Overlapping Attention for Optical Flow Estimation arXiv:2203.11335
Optical Flow EstimationonKITTI 2015
Fl-all· 2022-03-21
4.79
best: 11.22 (FastFlowNet-ft)
Global Matching with Overlapping Attention for Optical Flow Estimation arXiv:2203.11335
Optical Flow EstimationonKITTI 2015
Fl-fg· 2022-03-21
6.84
best: 7.75 (LiteFlowNet3)
Global Matching with Overlapping Attention for Optical Flow Estimation arXiv:2203.11335
Optical Flow EstimationonKITTI 2015 (train)
EPE· 2022-03-21
4.24
best: 13.17 (HD3)
Global Matching with Overlapping Attention for Optical Flow Estimation arXiv:2203.11335
Optical Flow EstimationonKITTI 2015 (train)
F1-all· 2022-03-21
15.4
best: 33.7 (PWC-Net)
Global Matching with Overlapping Attention for Optical Flow Estimation arXiv:2203.11335