DEVA (Mask2Former)

Reported on 3 benchmarks across 1 task · 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

Video SegmentationonBURST-val
OWTA (all)· uses extra data· 2023-09-07
69.9
SOTA
Tracking Anything with Decoupled Video Segmentation arXiv:2309.03903
Video SegmentationonBURST-val
OWTA (com)· uses extra data· 2023-09-07
75.2
SOTA
Tracking Anything with Decoupled Video Segmentation arXiv:2309.03903
Video SegmentationonBURST-val
OWTA (unc)· uses extra data· 2023-09-07
41.5
best: 50.5 (DEVA (EntitySeg))
Tracking Anything with Decoupled Video Segmentation arXiv:2309.03903