Share your thoughts, 1 month free Claude Pro on usSee more

Home/Benchmarks

Audio-Visual Semantic Segmentation on AVS-Bench AVSS

26.52mIoU

Crab

Updated 1mo ago

Evaluation Results

Method	Links
Crab 2025.03		26.52	32.1

SOTA Paper

Crab

Crab: A Unified Audio-Visual Scene Understanding Model with Explicit Cooperation

Dataset

AVS-Bench

Follow for update

@wizwand_team Discord

Related Benchmarks

Audio-Visual Segmentation on AVS-Bench S4 (test)Audio-Visual Segmentation on AVS-Bench AVSS (test)Referring Audio-Visual Segmentation on Ref-AVS Seen (test)Referring Audio-Visual Segmentation on Ref-AVS Unseen (test)Referring Audio-Visual Segmentation on Ref-AVS (test)

© 2026 wizwand

Blog Contact Changelog Swarm

Privacy Policy Terms of Service FAQs Swarm Docs