Map Segmentation on nuScenes (val)
70.6mIoUDDP
Evaluation results
| Method | Links | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| DDPModality=Camera + LiDAR, Steps=32023.03 | 70.6 | 89.4 | — | 63.8 | — | — | — | — | 62.6 | 69.8 | 75 | 63 | — | — | — | — | — | |
| DDPModality=Camera + LiDAR, Steps=12023.03 | 70.3 | 89.3 | — | 63.5 | — | — | — | — | 62.3 | 69.5 | 74.8 | 62.5 | — | — | — | — | — | |
| MapFusionVenue=IF’252026.06 | 68.9 | — | — | — | — | — | — | — | — | 69.6 | — | 63 | — | 88.9 | 74 | 56.5 | 61.5 | |
| DSVT-PEncoder=Pillar, Deeper BEV backbone=true2023.01 | 68 | 87.6 | — | — | 67.2 | 72.7 | 59.7 | 62.7 | 58.2 | — | — | — | — | — | — | — | — | |
| X-AlignModality=Camera + LiDAR2023.03 | 65.7 | 86.8 | — | 57.1 | — | — | — | — | 58.2 | 65.2 | 70 | 58.3 | — | — | — | — | — | |
| GraphBEV++Lifting=LSS, Venue=-2026.06 | 63.8 | — | — | — | — | — | — | — | — | 61.5 | — | 53.6 | — | 86.8 | 69.7 | 58.1 | 53.1 | |
| GraphBEVVenue=ECCV’242026.06 | 63.3 | — | — | — | — | — | — | — | — | 60.9 | — | 53.1 | — | 86.3 | 69.1 | 57.5 | 53.1 | |
| BEVFusionModality=Camera + LiDAR2023.03 | 62.7 | 85.5 | — | 57 | — | — | — | — | 53.7 | 60.5 | 67.6 | 52 | — | — | — | — | — | |
| BEVFusionVenue=ICRA’232026.06 | 62.7 | — | — | — | — | — | — | — | — | 60.5 | — | 52 | — | 85.5 | 67.6 | 57 | 53.7 | |
| BridgeTAVenue=ArXiv’252026.06 | 60.8 | — | — | — | — | — | — | — | — | 58.6 | — | 53.6 | — | 83.3 | 62.9 | 56.6 | 50.1 | |
| RGCVenue=WACV’242026.06 | 59.7 | — | — | — | — | — | — | — | — | 57.1 | — | 51.7 | — | 81.7 | 60.5 | 53.8 | 53.5 | |
| DDPModality=Camera, Steps=32023.03 | 59.4 | 83.6 | — | 51.4 | — | — | — | — | 49.2 | 58.3 | 61.8 | 52.3 | — | — | — | — | — | |
| DDPVenue=ICCV’232026.06 | 59.4 | — | — | — | — | — | — | — | — | 58.3 | — | 52.4 | — | 83.6 | 61.6 | 51.4 | 49.2 | |
| DDPModality=Camera, Steps=12023.03 | 59.3 | 83.2 | — | 51.1 | — | — | — | — | 48.9 | 58.5 | 61.6 | 52.4 | — | — | — | — | — | |
| MetaBEVVenue=ICCV’232026.06 | 59.3 | — | — | — | — | — | — | — | — | 56.7 | — | 50.8 | — | 83.3 | 61.4 | 55.5 | 48 | |
| X-AlignModality=Camera2023.03 | 58 | 82.4 | — | 53.8 | — | — | — | — | 47.4 | 55.6 | 59.3 | 49.6 | — | — | — | — | — | |
| X-AlignVenue=WACV’232026.06 | 58 | — | — | — | — | — | — | — | — | 55.6 | — | 49.6 | — | 82.4 | 59.3 | 53.8 | 47.4 | |
| MapPriorVenue=ICCV’232026.06 | 56.7 | — | — | — | — | — | — | — | — | 54.6 | — | 46.7 | — | 81.7 | 58.3 | 53.3 | 45.1 | |
| BEVFusionModality=Camera2023.03 | 56.6 | 81.7 | — | 50.7 | — | — | — | — | 46.4 | 54.8 | 58.4 | 47.4 | — | — | — | — | — | |
| DSVT-PEncoder=Pillar, Deeper BEV backbone=false2023.01 | 51.6 | 79.7 | — | — | 51.8 | 61.1 | 38.2 | 33.8 | 45.3 | — | — | — | — | — | — | — | — | |
| PointPaintingModality=Camera + LiDAR2023.03 | 49.1 | 75.9 | — | 34.5 | — | — | — | — | 41.9 | 48.5 | 57.1 | 36.9 | — | — | — | — | — | |
| PointPaintingVenue=CVPR’202026.06 | 49.1 | — | — | — | — | — | — | — | — | 48.5 | — | 36.9 | — | 75.9 | 57.1 | 34.5 | 41.9 | |
| MVPModality=Camera + LiDAR2023.03 | 49 | 76.1 | — | 33 | — | — | — | — | 42.2 | 48.7 | 57 | 36.9 | — | — | — | — | — | |
| MVPVenue=NeurIPS’212026.06 | 49 | — | — | — | — | — | — | — | — | 48.7 | — | 36.9 | — | 76.1 | 57 | 33 | 42.2 | |
| 3D SpConvEncoder=3D SpConv2023.01 | 48.6 | 75.6 | — | — | 48.4 | 57.5 | 36.5 | 31.7 | 41.9 | — | — | — | — | — | — | — | — | |
| CenterPointVenue=CVPR’212026.06 | 48.6 | — | — | — | — | — | — | — | — | 48.4 | — | 36.5 | — | 75.6 | 57.5 | 31.7 | 41.9 | |
| LSSModality=Camera2023.03 | 44.4 | 75.4 | — | 39.1 | — | — | — | — | 36.5 | 38.8 | 46.3 | 30.3 | — | — | — | — | — | |
| LSSVenue=ECCV’202026.06 | 44.4 | — | — | — | — | — | — | — | — | 38.8 | — | 30.3 | — | 75.4 | 46.3 | 39.1 | 36.5 | |
| 2D ConvEncoder=2D Conv2023.01 | 43.8 | 72 | — | — | 43.1 | 53.1 | 29.7 | 27.7 | 37.5 | — | — | — | — | — | — | — | — | |
| PointPillarsVenue=CVPR’192026.06 | 43.8 | — | — | — | — | — | — | — | — | 43.1 | — | 29.7 | — | 72 | 53.1 | 27.7 | 37.5 | |
| OFTModality=Camera2023.03 | 42.1 | 74 | — | 35.9 | — | — | — | — | 33.9 | 35.3 | 45.9 | 27.5 | — | — | — | — | — | |
| M2BEVVenue=ArXiv’222026.06 | 40.5 | — | — | — | — | — | — | — | — | — | — | — | — | 77.2 | — | — | — | |
| CVTModality=Camera2023.03 | 40.2 | 74.3 | — | 35 | — | — | — | — | 29.4 | 36.8 | 39.9 | 25.8 | — | — | — | — | — | |
| CVTVenue=CVPR’222026.06 | 40.2 | — | — | — | — | — | — | — | — | 36.8 | — | 25.8 | — | 74.3 | 39.9 | 35 | 29.4 | |
| NRSegVenue=TIP’262026.06 | 24.6 | — | — | — | — | — | — | — | — | 16.9 | — | 12.1 | — | 59.1 | 21.9 | 16.8 | 21 | |
| BEVCarSensor Modality=Camera-radar, Image Resolution=(448, 896)2026.02 | — | — | 45.3 | — | — | — | — | — | — | — | — | — | — | 83.3 | — | — | — | |
| BEVDepth-SegResolution=200 x 200, Head=U-Net-like2022.11 | — | 82.7 | 46.4 | 45 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFormerResolution=200 x 2002022.11 | — | 77.5 | 23.9 | 46.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFormerSensor Modality=Camera-only, Image Resolution=(448, 800)2026.02 | — | — | 25.7 | — | — | — | — | — | — | — | — | — | — | 80.1 | — | — | — | |
| BEVGen2024.06 | — | — | — | 5.9 | — | — | — | — | — | — | — | — | 50.1 | — | — | — | — | |
| BEVGuideSensor Modality=Camera-radar, Image Resolution=(448, 800)2026.02 | — | — | 44.2 | — | — | — | — | — | — | — | — | — | — | 76.7 | — | — | — | |
| FIERYResolution=200 x 2002022.11 | — | — | — | 38.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GaussianBeVSensor Modality=Camera-only, Image Resolution=(448, 800)2026.02 | — | — | 47.4 | — | — | — | — | — | — | — | — | — | — | 82.6 | — | — | — | |
| GaussianCaRSensor Modality=Camera-radar, Image Resolution=(448, 800)2026.02 | — | — | 50.1 | — | — | — | — | — | — | — | — | — | — | 82.9 | — | — | — | |
| LSSResolution=200 x 2002022.11 | — | 72.9 | 20 | 32.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LSSSensor Modality=Camera-only, Image Resolution=(448, 800)2026.02 | — | — | 20 | — | — | — | — | — | — | — | — | — | — | 72.9 | — | — | — | |
| M2BEVModality=Camera2023.03 | — | 77.2 | — | — | — | — | — | — | 40.5 | — | — | — | — | — | — | — | — | |
| M²BEVResolution=200 x 2002022.11 | — | 75.9 | 38 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MagicD.2024.06 | — | — | — | 29.5 | — | — | — | — | — | — | — | — | 58.6 | — | — | — | — | |
| MatrixVTResolution=200 x 200, Head=U-Net-like2022.11 | — | 83.5 | 44.8 | 46.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OracleModel=BEVFusion (single-frame)2024.06 | — | — | — | 34.6 | — | — | — | — | — | — | — | — | 72.2 | — | — | — | — | |
| SimGen2024.06 | — | — | — | 31.2 | — | — | — | — | — | — | — | — | 62.9 | — | — | — | — | |
| SimGen-nuScConditioning=nuScenes-style2024.06 | — | — | — | 29.9 | — | — | — | — | — | — | — | — | 60.6 | — | — | — | — | |
| Simple-BEV++Sensor Modality=Camera-radar, Image Resolution=(448, 896)2026.02 | — | — | 40.4 | — | — | — | — | — | — | — | — | — | — | 81.2 | — | — | — | |
| UniADModality=C, Pretrain=None2026.07 | — | — | 31.3 | — | — | — | — | 13.8 | — | — | — | — | — | 69.1 | — | — | 25.7 | |
| UniADModality=C, Pretrain=ViDAR [23]2026.07 | — | — | 35.2 | — | — | — | — | 17.1 | — | — | — | — | — | 73.9 | — | — | 29.1 | |
| UniAD-CRISPModality=C+R, Pretrain=None2026.07 | — | — | 32.4 | — | — | — | — | 15.1 | — | — | — | — | — | 69.8 | — | — | 26.7 | |
| UniAD-CRISPModality=C+R, Pretrain=CRISP2026.07 | — | — | 35.4 | — | — | — | — | 16.6 | — | — | — | — | — | 72.7 | — | — | 29.3 | |
| UVTRBackbone=ConvNeXt-S2024.11 | — | — | 15 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UVTR+UniPADBackbone=ConvNeXt-S2024.11 | — | — | 16.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UVTR+VisionPADBackbone=ConvNeXt-S2024.11 | — | — | 20.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — |