DEVA (ReferFormer)

Reported on 4 benchmarks across 2 tasks · 1 paper

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

Computer Vision4 results

Instance SegmentationonRefer-YouTube-VOS (2021 public validation)
J&F· uses extra data· 2023-09-07
66
best: 73.9 (MPG-SAM 2)
Tracking Anything with Decoupled Video Segmentation arXiv:2309.03903
Instance SegmentationonDAVIS 2017 (val)
J&F 1st frame· uses extra data· 2023-09-07
66.3
best: 72.5 (UNINEXT-H)
Tracking Anything with Decoupled Video Segmentation arXiv:2309.03903
Referring Expression SegmentationonRefer-YouTube-VOS (2021 public validation)
J&F· uses extra data· 2023-09-07
66
best: 73.9 (MPG-SAM 2)
Tracking Anything with Decoupled Video Segmentation arXiv:2309.03903
Referring Expression SegmentationonDAVIS 2017 (val)
J&F 1st frame· uses extra data· 2023-09-07
66.3
best: 72.5 (UNINEXT-H)
Tracking Anything with Decoupled Video Segmentation arXiv:2309.03903