Domain Generalization On Gta5 To Cityscapes
평가 지표
mIoU
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | mIoU | Paper Title | Repository |
|---|---|---|---|
| Self-adaptation (ResNet - 101) | 46.99 | Semantic Self-adaptation: Enhancing Generalization with a Single Sample | |
| Rein | 66.4 | Stronger, Fewer, & Superior: Harnessing Vision Foundation Models for Domain Generalized Semantic Segmentation | |
| GtA-SFDA Source-Only (DeepLabv2-ResNet101) | 43.5 | Generalize then Adapt: Source-Free Domain Adaptive Semantic Segmentation | |
| VLTSeg (EVA02-CLIP-L) | 65.6 | Strong but simple: A Baseline for Domain Generalized Dense Perception by CLIP-based Transfer Learning | |
| DIFF | 58.01 | Diffusion Features to Bridge Domain Gap for Semantic Segmentation | |
| tqdm (EVA02-CLIP-L) | 68.88 | Textual Query-Driven Mask Transformer for Domain Generalized Segmentation | |
| CMFormer | 55.31 | Learning Content-enhanced Mask Transformer for Domain Generalized Urban-Scene Segmentation |
0 of 7 row(s) selected.