Semantic Segmentation on NYU v2 68 (test)
63.6mIoUAbs. SOTA
Evaluation results
| Method | Links | |
|---|---|---|
| Abs. SOTAReference=SOTA from [72]2025.03 | 63.6 | |
| DINOv2 + DA(Norm)Adapter=Norm2025.03 | 55.1 | |
| DINOv2 + DA(SDP)Adapter=SDP2025.03 | 54.8 | |
| DINOv2 + DA(No Perlin)Adapter=No Perlin2025.03 | 54.6 | |
| DINOv3 + DA(SDP)Adapter=SDP2025.03 | 54.5 | |
| DepthAnythingV2 + DinoV2Modality=Depth Aware RGB2025.03 | 53.5 | |
| DINOv32025.03 | 53 | |
| DINOv22025.03 | 52.7 | |
| DepthAnythingV2Modality=Depth Aware RGB2025.03 | 52 | |
| DuneModality=Depth Aware RGB2025.03 | 49 | |
| FiT3DModality=Depth Aware RGB2025.03 | 46.6 | |
| OmnivoreModality=Multi-Modal2025.03 | 43.5 | |
| Omnivore + DepthModality=Multi-Modal2025.03 | 42.3 | |
| DFormerV1-preModality=RGBD2025.03 | 42.2 | |
| MASt3R (ViT-L)Modality=Depth Aware RGB2025.03 | 38.5 | |
| EVA-02 + DA(SDP)Adapter=SDP2025.03 | 38.1 | |
| MultiMAE + DepthModality=Multi-Modal2025.03 | 36.7 | |
| DFormerV2-preModality=RGBD2025.03 | 33.5 | |
| OmniDC-preModality=RGBD2025.03 | 31.6 | |
| DMD3C-preModality=RGBD2025.03 | 31 | |
| EVA-022025.03 | 30.1 | |
| MultiMAEModality=Multi-Modal2025.03 | 29.7 | |
| DA(No pre-training) + DA(SDP)Adapter=SDP2025.03 | 29.6 | |
| CroCo V2Modality=Depth Aware RGB2025.03 | 24.3 | |
| DA(No pre-training)Pre-training=None2025.03 | 22.4 |