Semantic Segmentation on nuScenes (test)
85.1mIoUDITR
Evaluation Results
| Method | Links | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| DITRModality=2D-3D fusion2025.03 | 85.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Flash3Dconstraint=Same memory costs as PTv32024.12 | 83.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3 + M3NetYear=20242023.12 | 83.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Flash3Dconstraint=Same parameters as PTv32024.12 | 83.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3 + PPTYear=20242023.12 | 83 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3Training Protocol=Multi-dataset joint training (PPT [92])2023.12 | 83 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3 + PPTYear=20242023.08 | 83 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PPT + PTv3Params.=46.3M, Training Protocol=supervised pre-training (fine-tuned)2023.08 | 83 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3/PPT#Params=46.3M, Joint training=true2026.04 | 83 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3Year=20242023.12 | 82.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3Training Protocol=Trained from scratch2023.12 | 82.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3Year=20242023.08 | 82.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3Params.=46.2M2023.08 | 82.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv32024.12 | 82.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3Modality=3D-only2025.03 | 82.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv3#Params=46.1M, Joint training=false2026.04 | 82.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv2Year=20222023.12 | 82.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv2Training Protocol=Trained from scratch2023.12 | 82.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv2Year=20222023.08 | 82.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv22024.12 | 82.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv2Modality=3D-only2025.03 | 82.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PTv2#Params=12.8M, Joint training=false2026.04 | 82.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Volt-B#Params=94.3M, Joint training=true2026.04 | 82.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SphereFormerInput=L2023.03 | 81.9 | 91.7 | 83.3 | 39.2 | 94.7 | 92.5 | 77.5 | 84.2 | 84.4 | 79.1 | 88.4 | 78.3 | 97.9 | 69 | 81.5 | 77.2 | 93.4 | 90.2 | — | |
| SphereFormerInput Modality=LiDAR2023.03 | 81.9 | 91.7 | 83.3 | 39.2 | 94.7 | 92.5 | 77.5 | 84.2 | 84.4 | 79.1 | 88.4 | 78.3 | 97.9 | 69 | 81.5 | 77.2 | 93.4 | 90.2 | — | |
| SphereFormerYear=20232023.12 | 81.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SphereFormerTraining Protocol=Trained from scratch2023.12 | 81.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SphereFormerYear=20232023.08 | 81.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SphereFormerParams.=32.3M2023.08 | 81.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SphereFormer2024.12 | 81.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SphereFormerModality=3D-only2025.03 | 81.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SphereFormer#Params=32.3M, Joint training=false2026.04 | 81.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MSeg3DModality=2D-3D fusion2025.03 | 81.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| 2DPASSInput=L2022.07 | 80.8 | 90.1 | 81.7 | 55.3 | 92 | 91.8 | 73.3 | 86.5 | 78.5 | 72.5 | 84.7 | 75.5 | 97.6 | 69.1 | 79.9 | 75.5 | 90.2 | 88 | 44 | |
| 2DPASSInput=L2023.03 | 80.8 | 90.1 | 81.7 | 55.3 | 92 | 91.8 | 73.3 | 86.5 | 78.5 | 72.5 | 84.7 | 75.5 | 97.6 | 69.1 | 79.9 | 75.5 | 90.2 | 88 | — | |
| 2DPASSInput Modality=LiDAR2023.03 | 80.8 | 90.1 | 81.7 | 55.3 | 92 | 91.8 | 73.3 | 86.5 | 78.5 | 72.5 | 84.7 | 75.5 | 97.6 | 69.1 | 79.9 | 75.5 | 90.2 | 88 | — | |
| 2DPASSYear=20222023.12 | 80.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| 2DPASSTraining Protocol=Trained from scratch2023.12 | 80.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| 2DPASSYear=20222023.08 | 80.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| 2DPASSModality=2D-3D fusion2025.03 | 80.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| 4D-FormerModality=2D-3D fusion2025.03 | 80.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RangeFormerYear=20232023.12 | 80.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RangeFormerYear=20232023.08 | 80.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| 2D3DNetInput=L+C2022.07 | 80 | 90.1 | 83 | 59.4 | 88 | 85.1 | 63.7 | 84.4 | 82 | 76 | 84.8 | 71.9 | 96.9 | 67.4 | 79.8 | 76 | 92.1 | 89.2 | — | |
| 2D3DNetInput=L+C2023.03 | 80 | 90.1 | 83 | 59.4 | 88 | 85.1 | 63.7 | 84.4 | 82 | 76 | 84.8 | 71.9 | 96.9 | 67.4 | 79.8 | 76 | 92.1 | 89.2 | — | |
| 2D3DNetInput Modality=LiDAR+Camera2023.03 | 80 | 90.1 | 83 | 59.4 | 88 | 85.1 | 63.7 | 84.4 | 82 | 76 | 84.8 | 71.9 | 96.9 | 67.4 | 79.8 | 76 | 92.1 | 89.2 | — | |
| LCPSModality=2D-3D fusion2025.03 | 78.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| (AF)2-S3NetInput=L2022.07 | 78.3 | 88.5 | 78.9 | 52.2 | 89.9 | 84.2 | 77.4 | 74.3 | 77.3 | 72 | 83.9 | 73.8 | 97.1 | 66.5 | 77.5 | 74 | 87.7 | 86.8 | 270 | |
| (AF)2-S3NetInput=L2023.03 | 78.3 | 88.5 | 78.9 | 52.2 | 89.9 | 84.2 | 77.4 | 74.3 | 77.3 | 72 | 83.9 | 73.8 | 97.1 | 66.5 | 77.5 | 74 | 87.7 | 86.8 | — | |
| (AF)2-S3NetInput Modality=LiDAR2023.03 | 78.3 | 88.5 | 78.9 | 52.2 | 89.9 | 84.2 | 77.4 | 74.3 | 77.3 | 72 | 83.9 | 73.8 | 97.1 | 66.5 | 77.5 | 74 | 87.7 | 86.8 | — | |
| AF2S3NetTraining Protocol=Trained from scratch2023.12 | 78 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| AF2S3Net2024.12 | 78 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BaselineInput=L2022.07 | 77.6 | 88.5 | 80.8 | 37.9 | 92.7 | 90.5 | 65.4 | 77.6 | 71.5 | 70.9 | 83.1 | 75.3 | 97 | 69.3 | 78.1 | 75.6 | 89.1 | 86.8 | 44 | |
| SPVCNNInput=L2022.07 | 77.4 | 89.7 | 80 | 30 | 91.9 | 90.8 | 64.7 | 79 | 75.6 | 70.9 | 81 | 74.6 | 97.4 | 69.2 | 80 | 76.1 | 89.3 | 87.1 | 63 | |
| SPVCNNInput=L2023.03 | 77.4 | 89.7 | 80 | 30 | 91.9 | 90.8 | 64.7 | 79 | 75.6 | 70.9 | 81 | 74.6 | 97.4 | 69.2 | 80 | 76.1 | 89.3 | 87.1 | — | |
| SPVCNNInput Modality=LiDAR2023.03 | 77.4 | 89.7 | 80 | 30 | 91.9 | 90.8 | 64.7 | 79 | 75.6 | 70.9 | 81 | 74.6 | 97.4 | 69.2 | 80 | 76.1 | 89.3 | 87.1 | — | |
| SPVNASParams.=10.8M2023.08 | 77.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| AMVNetInput=L2022.07 | 77.3 | 90.1 | 80.6 | 32 | 81.7 | 88.9 | 67.1 | 84.3 | 76.1 | 73.5 | 84.9 | 67.3 | 97.5 | 67.4 | 79.4 | 75.5 | 91.5 | 88.7 | 85 | |
| AMVNetInput=L2023.03 | 77.3 | 90.1 | 80.6 | 32 | 81.7 | 88.9 | 67.1 | 84.3 | 76.1 | 73.5 | 84.9 | 67.3 | 97.5 | 67.4 | 79.4 | 75.5 | 91.5 | 88.7 | — | |
| AMVNetInput Modality=LiDAR2023.03 | 77.3 | 90.1 | 80.6 | 32 | 81.7 | 88.9 | 67.1 | 84.3 | 76.1 | 73.5 | 84.9 | 67.3 | 97.5 | 67.4 | 79.4 | 75.5 | 91.5 | 88.7 | — | |
| Cylinder3DInput=L2022.07 | 77.2 | 89.9 | 82.8 | 29.8 | 84.3 | 89.4 | 63 | 79.3 | 77.2 | 73.4 | 84.6 | 69.1 | 97.7 | 70.2 | 80.3 | 75.5 | 90.4 | 87.6 | 63 | |
| Cylinder3DInput=L2023.03 | 77.2 | 89.9 | 82.8 | 29.8 | 84.3 | 89.4 | 63 | 79.3 | 77.2 | 73.4 | 84.6 | 69.1 | 97.7 | 70.2 | 80.3 | 75.5 | 90.4 | 87.6 | — | |
| Cylinder3DInput Modality=LiDAR2023.03 | 77.2 | 89.9 | 82.8 | 29.8 | 84.3 | 89.4 | 63 | 79.3 | 77.2 | 73.4 | 84.6 | 69.1 | 97.7 | 70.2 | 80.3 | 75.5 | 90.4 | 87.6 | — | |
| Cylinder3DYear=20212023.12 | 77.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Cylinder3DTraining Protocol=Trained from scratch2023.12 | 77.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Cylinder3DYear=20212023.08 | 77.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Cylinder3DParams.=26.1M2023.08 | 77.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Cylinder3D2024.12 | 77.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Cylinder3DModality=3D-only2025.03 | 77.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Cylinder3D#Params=26.1M, Joint training=false2026.04 | 77.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PMFInput=L+C2022.07 | 77 | 89 | 82 | 40 | 81 | 88 | 64 | 79 | 80 | 76 | 81 | 67 | 97 | 68 | 78 | 74 | 90 | 88 | 125 | |
| PMFInput=L+C2023.03 | 77 | 89 | 82 | 40 | 81 | 88 | 64 | 79 | 80 | 76 | 81 | 67 | 97 | 68 | 78 | 74 | 90 | 88 | — | |
| PMFInput Modality=LiDAR+Camera2023.03 | 77 | 89 | 82 | 40 | 81 | 88 | 64 | 79 | 80 | 76 | 81 | 67 | 97 | 68 | 78 | 74 | 90 | 88 | — | |
| PVKDYear=20222023.12 | 76 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PVKDYear=20222023.08 | 76 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| JS3C-NetInput=L2022.07 | 73.6 | 88.1 | 80.1 | 26.2 | 87.8 | 84.5 | 55.2 | 72.6 | 71.3 | 66.3 | 76.8 | 71.2 | 96.8 | 64.5 | 76.9 | 74.1 | 87.5 | 86.1 | — | |
| JS3C-NetInput=L2023.03 | 73.6 | 88.1 | 80.1 | 26.2 | 87.8 | 84.5 | 55.2 | 72.6 | 71.3 | 66.3 | 76.8 | 71.2 | 96.8 | 64.5 | 76.9 | 74.1 | 87.5 | 86.1 | — | |
| JS3C-NetInput Modality=LiDAR2023.03 | 73.6 | 88.1 | 80.1 | 26.2 | 87.8 | 84.5 | 55.2 | 72.6 | 71.3 | 66.3 | 76.8 | 71.2 | 96.8 | 64.5 | 76.9 | 74.1 | 87.5 | 86.1 | — | |
| PolarNetInput=L2022.07 | 69.4 | 87.4 | 72.2 | 16.8 | 77 | 86.5 | 51.1 | 69.7 | 64.8 | 54.1 | 69.7 | 63.5 | 96.6 | 67.1 | 77.7 | 72.1 | 87.1 | 84.5 | — | |
| PolarNetInput=L2023.03 | 69.4 | 87.4 | 72.2 | 16.8 | 77 | 86.5 | 51.1 | 69.7 | 64.8 | 54.1 | 69.7 | 63.5 | 96.6 | 67.1 | 77.7 | 72.1 | 87.1 | 84.5 | — | |
| PolarNetInput Modality=LiDAR2023.03 | 69.4 | 87.4 | 72.2 | 16.8 | 77 | 86.5 | 51.1 | 69.7 | 64.8 | 54.1 | 69.7 | 63.5 | 96.6 | 67.1 | 77.7 | 72.1 | 87.1 | 84.5 | — |