Semantic Segmentation on DSEC-Semantic v1 (test)
64.93mIoUScaleEvent
Evaluation results
| Method | Links | |
|---|---|---|
| ScaleEventBackbone=ViT-B/16, Protocol=Full2026.03 | 64.93 | |
| ScaleEventBackbone=ViT-B/16, Protocol=20%2026.03 | 64.15 | |
| ScaleEventBackbone=ViT-B/16, Protocol=10%2026.03 | 63.88 | |
| ScaleEventBackbone=ViT-B/16, Protocol=5%2026.03 | 62.82 | |
| ScaleEventBackbone=ViT-B/16, Protocol=LP2026.03 | 58.42 | |
| OpenESSBackbone=E2VID, Protocol=Full2026.03 | 57.21 | |
| FC-CLIPBackbone=ConvNeXt-L, Protocol=Full2026.03 | 55.67 | |
| MaskCLIPBackbone=ViT-B/16, Protocol=Full2026.03 | 55.01 | |
| ScaleEventBackbone=ViT-B/16, Protocol=1%2026.03 | 54.37 | |
| OpenESSBackbone=E2VID, Protocol=20%2026.03 | 48.28 | |
| FC-CLIPBackbone=ConvNeXt-L, Protocol=20%2026.03 | 47.77 | |
| OpenESSBackbone=E2VID, Protocol=10%2026.03 | 46.25 | |
| OpenESSBackbone=E2VID, Protocol=5%2026.03 | 44.97 | |
| OpenESSBackbone=E2VID, Protocol=LP2026.03 | 44.26 | |
| FC-CLIPBackbone=ConvNeXt-L, Protocol=10%2026.03 | 44.09 | |
| FC-CLIPBackbone=ConvNeXt-L, Protocol=5%2026.03 | 43.71 | |
| FC-CLIPBackbone=ConvNeXt-L, Protocol=LP2026.03 | 43 | |
| MaskCLIPBackbone=ViT-B/16, Protocol=20%2026.03 | 42.4 | |
| OpenESSBackbone=E2VID, Protocol=1%2026.03 | 41.41 | |
| FC-CLIPBackbone=ConvNeXt-L, Protocol=1%2026.03 | 39.12 | |
| MaskCLIPBackbone=ViT-B/16, Protocol=10%2026.03 | 38.83 | |
| MaskCLIPBackbone=ViT-B/16, Protocol=5%2026.03 | 37.03 | |
| MaskCLIPBackbone=ViT-B/16, Protocol=1%2026.03 | 33.89 | |
| MaskCLIPBackbone=ViT-B/16, Protocol=LP2026.03 | 33.08 |