3D Object Detection on KITTI (val) (Class AP3D Breakdown)
96.2mAP3D - Car (Easy)Motal
Evaluation results
| Method | Links | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| MotalTrain Set=Waymo2026.02 | 96.2 | 87.6 | 85.8 | 37.9 | 33.4 | 31.1 | 56.3 | 37.8 | 35.5 | — | — | — | — | |
| VirConv-TSpeed (FPS)=10.22026.01 | 94.98 | 89.96 | 88.13 | 73.32 | 66.93 | 60.38 | 90.04 | 73.9 | 69.06 | — | — | — | — | |
| SPLTrain Set=KITTI2026.02 | 93.3 | 83.1 | 75.9 | 46.1 | 40.5 | 34 | 67.1 | 41.9 | 36.7 | — | — | — | — | |
| LCF3D-StereoSpeed (FPS)=10.1, RGB branch detector=Faster RCNN (FR), LiDAR branch detector=PV-RCNN (PV)2026.01 | 92.95 | 86.09 | 83.32 | 73.87 | 67.4 | 62.67 | 91.01 | 77.25 | 72.01 | — | — | — | — | |
| LCF3D-SingleSpeed (FPS)=10.5, RGB branch detector=Faster RCNN (FR), LiDAR branch detector=PV-RCNN (PV)2026.01 | 92.44 | 85.99 | 83.54 | 73.43 | 68.18 | 63.56 | 89.83 | 77.27 | 72.17 | — | — | — | — | |
| Voxel-RCNNAnnotation Rate=100%2026.02 | 92.3 | 84.9 | 82.6 | 69.6 | 63 | 58.6 | 88.7 | 72.5 | 68.2 | — | — | — | — | |
| LoGoNet2026.01 | 92.04 | 85.04 | 84.31 | 70.2 | 63.72 | 59.46 | 91.74 | 75.35 | 72.42 | — | — | — | — | |
| CoIn++Annotation Rate=2%2026.02 | 92 | 79.5 | 71.5 | 46.7 | 36.1 | 31.2 | 82 | 58.4 | 54.6 | — | — | — | — | |
| SPLAnnotation Rate=2%2026.02 | 91.8 | 82.2 | 79.4 | 69.5 | 63.2 | 56.7 | 91.8 | 73 | 68.5 | — | — | — | — | |
| SP3DAnnotation Rate=2%2026.02 | 91.3 | 80.5 | 74 | 67.4 | 58.7 | 50.9 | 92.5 | 73.1 | 68.3 | — | — | — | — | |
| CPDTrain Set=Waymo2026.02 | 90.9 | 81 | 79.8 | 17.1 | 15.2 | 14.2 | 11.1 | 7.3 | 6.5 | — | — | — | — | |
| CAT-DetSpeed (FPS)=10.22026.01 | 90.12 | 81.46 | 79.15 | 74.08 | 66.35 | 58.92 | 87.64 | 72.82 | 68.2 | — | — | — | — | |
| MLF-DET-VSpeed (FPS)=10.82026.01 | 89.7 | 87.31 | 79.34 | 71.15 | 68.5 | 61.72 | 86.05 | 72.14 | 65.42 | — | — | — | — | |
| CLOCs-PVCas2026.01 | 89.49 | 79.31 | 77.36 | 62.88 | 56.2 | 50.1 | 87.57 | 67.92 | 63.67 | — | — | — | — | |
| LCF3D-SingleSpeed (FPS)=30.4, RGB branch detector=Faster RCNN (FR), LiDAR branch detector=PointPillars (PP)2026.01 | 89.3 | 80.03 | 77.23 | 69.81 | 64.66 | 59.9 | 87.13 | 75.74 | 70.86 | — | — | — | — | |
| CoInAnnotation Rate=2%2026.02 | 89.1 | 70.2 | 55.6 | 50.8 | 45.2 | 39.6 | 80.2 | 52.3 | 48.6 | — | — | — | — | |
| Frustum PointPillarsSpeed (FPS)=14.32026.01 | 88.9 | 79.28 | 78.07 | 66.11 | 61.89 | 56.91 | 87.54 | 72.78 | 66.07 | — | — | — | — | |
| SECONDDetection Head=anchor-based region proposal network2023.03 | 88.61 | 78.62 | 77.22 | 56.55 | 52.98 | 47.73 | 80.58 | 67.15 | 63.1 | — | — | — | — | |
| SECOND2026.05 | 88.51 | 78.19 | 76.01 | 52.12 | 46.08 | 42.06 | 78.48 | 64.64 | 60.57 | 80.9 | 46.75 | 67.9 | 65.18 | |
| MVXNet2026.01 | 88.48 | 78.75 | 74.34 | 58.27 | 55.51 | 51.83 | 79.15 | 63.25 | 60.56 | — | — | — | — | |
| PointPillarsDetection Head=anchor-based region proposal network2023.03 | 88.46 | 77.28 | 74.65 | 57.75 | 52.29 | 47.9 | 80.04 | 62.61 | 59.52 | — | — | — | — | |
| VoxSeTDetection Head=anchor-based region proposal network2023.03 | 88.45 | 78.48 | 77.07 | 60.62 | 54.74 | 50.39 | 84.07 | 68.11 | 65.14 | — | — | — | — | |
| OcTrDetection Head=anchor-based region proposal network2023.03 | 88.43 | 78.57 | 77.16 | 61.49 | 57.17 | 52.35 | 85.29 | 70.44 | 66.17 | — | — | — | — | |
| PointPainting2026.01 | 88.38 | 77.74 | 76.76 | 69.38 | 61.67 | 54.58 | 85.21 | 71.62 | 66.98 | — | — | — | — | |
| LION2026.05 | 88.12 | 77.84 | 76.51 | 62.11 | 55.82 | 50.48 | 79.39 | 61.37 | 57.93 | 80.82 | 56.14 | 66.23 | 67.73 | |
| 3DHMT2026.05 | 87.95 | 78 | 76.63 | 62.6 | 57.55 | 51.61 | 82.2 | 67.74 | 63.96 | 80.86 | 57.25 | 71.3 | 69.8 | |
| VoTRDetection Head=anchor-based region proposal network2023.03 | 87.86 | 78.27 | 76.93 | — | — | — | — | — | — | — | — | — | — | |
| DSVT2026.05 | 87.78 | 77.21 | 74.93 | 58.56 | 52.12 | 47.46 | 82.5 | 65.08 | 62.81 | 79.97 | 52.71 | 70.13 | 67.61 | |
| PointRCNN2026.05 | 85.94 | 75.76 | 68.32 | 49.43 | 41.78 | 38.63 | 73.93 | 59.6 | 53.59 | 76.67 | 43.28 | 62.37 | 60.78 | |
| Frustum PointNetSpeed (FPS)=5.92026.01 | 83.76 | 70.92 | 63.65 | 70 | 61.32 | 53.59 | 77.15 | 56.49 | 53.37 | — | — | — | — | |
| Fore-Mamba3DBackbone=Mamba2026.02 | 79.5 | 82.2 | 90.3 | 57 | 62.2 | 67.8 | 66.3 | 69.5 | 86.4 | — | — | — | — | |
| CenterPoint2026.05 | 79.26 | 69.21 | 64.8 | 32.96 | 30.19 | 28.03 | 69.92 | 53.29 | 51.35 | 71.09 | 30.39 | 58.19 | 53.22 | |
| PointPillars2026.05 | 79.05 | 74.99 | 68.3 | 52.08 | 43.53 | 41.49 | 75.78 | 59.07 | 52.92 | 74.11 | 45.7 | 62.59 | 59.2 | |
| DGT-DetBackbone=Transformer2026.02 | 78.8 | 80.6 | 89.6 | — | — | — | 61 | 68.9 | 82.1 | — | — | — | — | |
| VoxelMambaBackbone=Mamba2026.02 | 78.1 | 80.8 | 89.1 | 53.7 | 59.7 | 66 | 64.8 | 69.1 | 84.2 | — | — | — | — | |
| LIONBackbone=Mamba2026.02 | 77.2 | 78.3 | 88.6 | 55.6 | 60.2 | 67.2 | 63.9 | 68.6 | 83 | — | — | — | — | |
| DSVTBackbone=Transformer2026.02 | 76.8 | 77.8 | 87.8 | 55.2 | 59.7 | 66.1 | 63.2 | 66.7 | 83.5 | — | — | — | — | |
| IA-SSDBackbone=MLP2026.02 | 75 | 80.1 | 88.3 | 35.6 | 39 | 46.5 | 55.7 | 61.9 | 78.4 | — | — | — | — | |
| Voxel-RCNNAnnotation Rate=2%2026.02 | 70.5 | 54.9 | 44.8 | 42.6 | 38.5 | 32.1 | 73.3 | 47.8 | 43.2 | — | — | — | — | |
| PointPillarsBackbone=MLP2026.02 | 68.3 | 75 | 79.1 | 41.5 | 43.5 | 52.1 | 52.9 | 59.1 | 75.8 | — | — | — | — | |
| OYSTERTrain Set=Waymo2026.02 | 65.3 | 54.8 | 43.6 | 3 | 3 | 3 | 1.7 | 1.8 | 1.9 | — | — | — | — | |
| LISOTrain Set=KITTI2026.02 | 62.4 | 53.7 | 45.6 | 13.4 | 10.8 | 8.2 | 20.4 | 13.7 | 10.3 | — | — | — | — | |
| VoxelNetBackbone=SpCNN2026.02 | 57.7 | 65.1 | 77.5 | 31.5 | 33.7 | 39.5 | 44.4 | 48.4 | 61.2 | — | — | — | — | |
| MODESTTrain Set=Waymo2026.02 | 47.6 | 33.4 | 30.6 | 1.3 | 2.2 | 2.3 | 0.1 | 0 | 0 | — | — | — | — | |
| OYSTERTrain Set=KITTI2026.02 | 43.7 | 34.5 | 31.2 | 0 | 0 | 0 | 0 | 0 | 0 | — | — | — | — |