Video Semantic Segmentation on NYU v2
49.1mIoURS-SSM
Evaluation results
| Method | Links | |
|---|---|---|
| RS-SSMBackbone=MiT-B5, GFLOPs=128.8, Params(M)=88.4, FPS=20.2, Input Resolution=480 × 8532026.03 | 49.1 | |
| RS-SSMBackbone=Swin-S, GFLOPs=110.3, Params(M)=55.9, FPS=24.2, Input Resolution=480 × 8532026.03 | 47 | |
| TV3SBackbone=MiT-B5, GFLOPs=137.0, Params(M)=85.6, FPS=14.0, Input Resolution=480 × 8532026.03 | 46.9 | |
| CFFM++Backbone=MiT-B5, GFLOPs=444.3, Params(M)=87.5, FPS=10.4, Input Resolution=480 × 8532026.03 | 46.7 | |
| MRCFABackbone=MiT-B5, GFLOPs=373.0, Params(M)=84.5, FPS=5.0, Input Resolution=480 × 8532026.03 | 46.7 | |
| SegformerBackbone=MiT-B5, GFLOPs=185.0, Params(M)=82.1, FPS=9.4, Input Resolution=480 × 8532026.03 | 46.1 | |
| RS-SSMBackbone=MiT-B2, GFLOPs=60.7, Params(M)=31.2, FPS=30.9, Input Resolution=480 × 8532026.03 | 45.9 | |
| TV3SBackbone=Swin-S, GFLOPs=94.1, Params(M)=53.1, FPS=19.5, Input Resolution=480 × 8532026.03 | 45.2 | |
| TV3SBackbone=MiT-B2, GFLOPs=53.9, Params(M)=28.3, FPS=21.9, Input Resolution=480 × 8532026.03 | 44.7 | |
| CFFM++Backbone=MiT-B2, GFLOPs=172.6, Params(M)=28.5, FPS=21.5, Input Resolution=480 × 8532026.03 | 43.9 | |
| MRCFABackbone=MiT-B2, GFLOPs=127.9, Params(M)=27.3, FPS=10.7, Input Resolution=480 × 8532026.03 | 43.5 | |
| Mask2FormerBackbone=Swin-S, GFLOPs=152.2, Params(M)=68.9, FPS=14.5, Input Resolution=480 × 8532026.03 | 43.1 | |
| RS-SSMBackbone=Swin-T, GFLOPs=73.5, Params(M)=34.6, FPS=26.7, Input Resolution=480 × 8532026.03 | 42.2 | |
| SegformerBackbone=MiT-B2, GFLOPs=100.8, Params(M)=24.8, FPS=16.2, Input Resolution=480 × 8532026.03 | 42.1 | |
| Mask2FormerBackbone=R50, GFLOPs=110.6, Params(M)=44.0, FPS=19.4, Input Resolution=480 × 8532026.03 | 41.7 | |
| MPVSSBackbone=Swin-S, GFLOPs=47.3, Params(M)=108.0, FPS=30.6, Input Resolution=480 × 8532026.03 | 41.7 | |
| RS-SSMBackbone=MiT-B1, GFLOPs=48.1, Params(M)=20.2, FPS=37.3, Input Resolution=480 × 8532026.03 | 41.4 | |
| Mask2FormerBackbone=R101, GFLOPs=141.3, Params(M)=63.0, FPS=16.9, Input Resolution=480 × 8532026.03 | 41.3 | |
| TV3SBackbone=Swin-T, GFLOPs=57.3, Params(M)=31.7, FPS=22.9, Input Resolution=480 × 8532026.03 | 40.5 | |
| MPVSSBackbone=R101, GFLOPs=45.1, Params(M)=103.1, FPS=32.3, Input Resolution=480 × 8532026.03 | 40.4 | |
| MPVSSBackbone=R50, GFLOPs=38.9, Params(M)=84.1, FPS=33.9, Input Resolution=480 × 8532026.03 | 40.1 | |
| TV3SBackbone=MiT-B1, GFLOPs=36.9, Params(M)=17.3, FPS=24.7, Input Resolution=480 × 8532026.03 | 39.7 | |
| CFFM++Backbone=MiT-B1, GFLOPs=114.3, Params(M)=16.5, FPS=27.6, Input Resolution=480 × 8532026.03 | 39.5 | |
| MRCFABackbone=MiT-B1, GFLOPs=77.5, Params(M)=16.2, FPS=40.1, Input Resolution=480 × 8532026.03 | 39.1 | |
| SegformerBackbone=MiT-B1, GFLOPs=26.6, Params(M)=13.8, FPS=58.7, Input Resolution=480 × 8532026.03 | 37.6 | |
| Mask2FormerBackbone=Swin-T, GFLOPs=114.4, Params(M)=47.4, FPS=17.1, Input Resolution=480 × 8532026.03 | 37.4 | |
| MPVSSBackbone=Swin-T, GFLOPs=39.7, Params(M)=114.0, FPS=32.8, Input Resolution=480 × 8532026.03 | 36.8 |