ReferEverything: Towards Segmenting Everything We Can Speak of in Videos
Open paper
Benchmark
Model
Rank
Results
referring-video-object-segmentation-on-mevis
REM Wan-14B
–
J&F: 60.3
referring-video-object-segmentation-on-refvos-davis
REM Wan-14B
–
J&F: 75.0
Rank counts only results with a code link.