Semantic Segmentation on BDD
68.08mIoUSoMA
Evaluation results
| Method | Links | |
|---|---|---|
| SoMABackbone=DINOv2-L, Trainable Parameters=4.9M, Multi-scale flip augmentation=true2024.12 | 68.08 | |
| SoMABackbone=DINOv2-L, Trainable Parameters=4.9M, Multi-scale flip augmentation=false2024.12 | 67.02 | |
| ReinBackbone=DINOv2-L, Trainable Parameters=3.0M, Re-implemented=true2024.12 | 66.53 | |
| FADABackbone=DINOv2-L, Trainable Parameters=111.7M2024.12 | 65.12 | |
| tqdmBackbone=EVA02-L, Trainable Parameters=304.2M2024.12 | 64.72 | |
| CMFormerBackbone=Swin-L, Trainable Parameters=196.0M2024.12 | 62.6 | |
| HGFormerBackbone=Swin-L, Trainable Parameters=196.0M2024.12 | 61.5 | |
| CoCoOp+CAKIBase Model=SEEM-Large2026.05 | 50.7 | |
| CoCoOpBase Model=SEEM-Large2026.05 | 49.5 | |
| DiffTPT + HisTPTModel Scale=Large2024.10 | 47.8 | |
| TPT + HisTPTModel Scale=Large2024.10 | 47.4 | |
| HisTPTModel Scale=Large2024.10 | 47.3 | |
| CoOp+CAKIBase Model=SEEM-Large2026.05 | 46.3 | |
| DiffTPTModel Scale=Large2024.10 | 45.7 | |
| TPTModel Scale=Large2024.10 | 45.2 | |
| CoOpBase Model=SEEM-Large2026.05 | 45.2 | |
| SEEMModel Scale=Large2024.10 | 44.6 | |
| SEEM-Large2026.05 | 44.6 | |
| CoOp+CAKIBase Model=SEEM-Tiny2026.05 | 42.6 | |
| DiffTPT + HisTPTModel Scale=Tiny2024.10 | 42.1 | |
| TPT + HisTPTModel Scale=Tiny2024.10 | 41.8 | |
| CoOpBase Model=SEEM-Tiny2026.05 | 41.6 | |
| HisTPTModel Scale=Tiny2024.10 | 41.2 | |
| DiffTPTModel Scale=Tiny2024.10 | 39.6 | |
| CoCoOp+CAKIBase Model=SEEM-Tiny2026.05 | 39.6 | |
| TPTModel Scale=Tiny2024.10 | 38.9 | |
| CoCoOpBase Model=SEEM-Tiny2026.05 | 38.3 | |
| SEEMModel Scale=Tiny2024.10 | 37.4 | |
| SEEM-Tiny2026.05 | 37.4 |