DVIS++(ViT-L, Online)

Reported on 5 benchmarks across 1 task · 1 paper · 5 SOTA

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

Computer Vision5 results

Video Instance SegmentationonYouTube-VIS validation
AP50· uses extra data· 2023-12-20
88.8
best: 89.3 (CAVIS(ViT-L, Online))
SOTA
DVIS++: Improved Decoupled Framework for Universal Video Segmentation arXiv:2312.13305
Video Instance SegmentationonYouTube-VIS validation
AP75· uses extra data· 2023-12-20
75.3
best: 76.2 (CAVIS(ViT-L, Online))
SOTA
DVIS++: Improved Decoupled Framework for Universal Video Segmentation arXiv:2312.13305
Video Instance SegmentationonYouTube-VIS validation
AR1· uses extra data· 2023-12-20
57.9
best: 58.3 (CAVIS(ViT-L, Online))
SOTA
DVIS++: Improved Decoupled Framework for Universal Video Segmentation arXiv:2312.13305
Video Instance SegmentationonYouTube-VIS validation
AR10· uses extra data· 2023-12-20
73.7
SOTA
DVIS++: Improved Decoupled Framework for Universal Video Segmentation arXiv:2312.13305
Video Instance SegmentationonYouTube-VIS validation
mask AP· uses extra data· 2023-12-20
67.7
best: 68.9 (CAVIS(ViT-L, Online))
SOTA
DVIS++: Improved Decoupled Framework for Universal Video Segmentation arXiv:2312.13305