SOTAVerified

Video Object Segmentation

Video object segmentation is a binary labeling problem aiming to separate foreground object(s) from the background region of a video.

For leaderboards please refer to the different subtasks.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–10 of 551 papers

TitleStatusHype
SeC: Advancing Complex Video Object Segmentation via Progressive Concept Construction—0
Memory-Augmented SAM2 for Training-Free Surgical Video Segmentation—0
MUVOD: A Novel Multi-view Video Object Segmentation Dataset and A Benchmark for 3D Segmentation—0
M^3-VOS: Multi-Phase, Multi-Transition, and Multi-Scenery Video Object SegmentationCode1
THU-Warwick Submission for EPIC-KITCHEN Challenge 2025: Semi-Supervised Video Object Segmentation—0
VideoMolmo: Spatio-Temporal Grounding Meets PointingCode2
InterRVOS: Interaction-aware Referring Video Object Segmentation—0
Omni-R1: Reinforcement Learning for Omnimodal Reasoning via Two-System CollaborationCode2
ThinkVideo: High-Quality Reasoning Video Segmentation with Chain of ThoughtsCode0
Long-RVOS: A Comprehensive Benchmark for Long-term Referring Video Object Segmentation—0
Show:102550
← PrevPage 1 of 56Next →
All datasetsDAVIS 2016DAVIS 2017 (val)YouTube-VOS 2018DAVIS-2017 (test-dev)YouTube-VOS 2019DAVIS 2017M^3-VOSFBMSYouTubeFBMS-59MOSESegTrack-v2

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1AOC-MF (val)Jaccard (Mean)81.7—Unverified
2ViTAE-T-StageJaccard (Mean)79.4—Unverified
3DINO (ViT-B/8, ImageNet retrain)J&F71.4—Unverified
4VOSwL (Mask+Language)mIoU59—Unverified
5UniTrackmIoU58.4—Unverified