DML-CSR

Reported on 6 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 Vision3 results

Scene ParsingonHelen
Mean F1· 2022-03-28
93.8
SOTA
Decoupled Multi-task Learning with Cyclical Self-Regulation for Face Parsing arXiv:2203.14448
Scene ParsingonCelebAMask-HQ
Mean F1· 2022-03-28
86.1
best: 89.56 (FaRL-B)
Decoupled Multi-task Learning with Cyclical Self-Regulation for Face Parsing arXiv:2203.14448
Scene ParsingonLaPa
Mean F1· 2022-03-28
92.4
best: 93.88 (FaRL-B)
Decoupled Multi-task Learning with Cyclical Self-Regulation for Face Parsing arXiv:2203.14448

Audio3 results

2D Semantic SegmentationonHelen
Mean F1· 2022-03-28
93.8
SOTA
Decoupled Multi-task Learning with Cyclical Self-Regulation for Face Parsing arXiv:2203.14448
2D Semantic SegmentationonCelebAMask-HQ
Mean F1· 2022-03-28
86.1
best: 89.56 (FaRL-B)
Decoupled Multi-task Learning with Cyclical Self-Regulation for Face Parsing arXiv:2203.14448
2D Semantic SegmentationonLaPa
Mean F1· 2022-03-28
92.4
best: 93.88 (FaRL-B)
Decoupled Multi-task Learning with Cyclical Self-Regulation for Face Parsing arXiv:2203.14448