Audio-Visual Segmentation on AVSBench MS3
54.5MJ ScoreSelM
Evaluation results
| Method | Links | ||
|---|---|---|---|
| SelMVisual Backbone=R50, Audio Backbone=VGGish, GPU Latency (ms)=25.3, Mobile Latency (ms)=1003.8, Params=117.6M2026.05 | 54.5 | 65.6 | |
| LightAVSegVisual Backbone=Sea, Audio Backbone=MNetV2, GPU Latency (ms)=15.9, Mobile Latency (ms)=163.4, Params=20.5M2026.05 | 50.4 | 62.6 | |
| AVSegFormerVisual Backbone=R50, Audio Backbone=VGGish, GPU Latency (ms)=29.0, Mobile Latency (ms)=1271.4, Params=151.1M2026.05 | 49.5 | 62.8 | |
| AVSBenchVisual Backbone=R50, Audio Backbone=VGGish, GPU Latency (ms)=21.2, Mobile Latency (ms)=753.5, Params=91.4M2026.05 | 47.9 | 57.8 | |
| SelMVisual Backbone=Sea, Audio Backbone=MNetV2, GPU Latency (ms)=18.7, Mobile Latency (ms)=308.6, Params=39.5M2026.05 | 45.7 | 57.6 | |
| AVSegFormerVisual Backbone=Sea, Audio Backbone=MNetV2, GPU Latency (ms)=22.2, Mobile Latency (ms)=432.6, Params=51.0M2026.05 | 40.7 | 50.7 | |
| AVSBenchVisual Backbone=Sea, Audio Backbone=MNetV2, GPU Latency (ms)=19.5, Mobile Latency (ms)=237.1, Params=30.2M2026.05 | 35.2 | 44.1 |