Video Semantic Segmentation on Cityscapes
75.6mIoUTV3S
Evaluation results
| Method | Links | |
|---|---|---|
| TV3SBackbone=MiT-B1, GFLOPs=83.6, Params(M)=17.3, FPS=25.1, Resolution=512 x 1024, Processing Mode=Frame-by-Frame2025.03 | 75.6 | |
| CFFMBackbone=MiT-B1, GFLOPs=158.7, Params(M)=15.4, FPS=11.7, Resolution=512 x 1024, Processing Mode=Frame-by-Frame2025.03 | 75.1 | |
| MRCFABackbone=MiT-B1, GFLOPs=145.0, Params(M)=14.9, FPS=13.0, Resolution=512 x 1024, Processing Mode=Frame-by-Frame2025.03 | 75.1 | |
| SegFormerBackbone=MiT-B1, Params(M)=13.8, FPS=46.8, Resolution=512 x 1024, Processing Mode=Frame-by-Frame2025.03 | 74.1 | |
| CFFMBackbone=MiT-B0, GFLOPs=80.7, Params(M)=4.6, FPS=15.8, Resolution=512 x 1024, Processing Mode=Frame-by-Frame2025.03 | 74 | |
| MRCFABackbone=MiT-B0, GFLOPs=77.5, Params(M)=4.2, FPS=16.6, Resolution=512 x 1024, Processing Mode=Frame-by-Frame2025.03 | 72.8 | |
| SegFormerBackbone=MiT-B0, Params(M)=3.7, FPS=58.5, Resolution=512 x 1024, Processing Mode=Frame-by-Frame2025.03 | 71.9 | |
| ETCBackbone=R18, GFLOPs=434.1, Resolution=512 x 1024, Processing Mode=Frame-by-Frame2025.03 | 71.1 |