Audio-Visual Segmentation on AVS-Bench AVSS (test)
29.8mIoUTPAVI
Evaluation results
| Method | Links | ||
|---|---|---|---|
| TPAVIVisual Backbone=PVT-v22025.03 | 29.8 | 35.2 | |
| CrabVisual Backbone=ViT-L/142025.03 | 26.59 | 32.1 | |
| AVSegFormerVisual Backbone=PVT-v22025.03 | 24.9 | 29.3 | |
| BAVSVisual Backbone=ResNet-502025.03 | 24.7 | 29.6 | |
| TPAVIVisual Backbone=ResNet-502025.03 | 20.2 | 25.2 |