DVIS++(VIT-L,Offline)

Reported on 3 benchmarks across 1 task · 1 paper · 3 SOTA

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

Computer Vision3 results

Video Instance SegmentationonOVIS validation
AP50· uses extra data· 2023-12-20
78.9
best: 83.8 (DVIS-DAQ(VIT-L, Offline))
SOTA
DVIS++: Improved Decoupled Framework for Universal Video Segmentation arXiv:2312.13305
Video Instance SegmentationonOVIS validation
AP75· uses extra data· 2023-12-20
58.5
best: 63.5 (CAVIS(VIT-L, Offline))
SOTA
DVIS++: Improved Decoupled Framework for Universal Video Segmentation arXiv:2312.13305
Video Instance SegmentationonOVIS validation
mask AP· uses extra data· 2023-12-20
53.4
best: 57.1 (DVIS-DAQ(VIT-L, Offline))
SOTA
DVIS++: Improved Decoupled Framework for Universal Video Segmentation arXiv:2312.13305