Bird's Eye View Object Detection on KITTI (val)
92.92AP BEV (Moderate)SA-SSD+EBM
Evaluation results
| Method | Links | ||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SA-SSD+EBMrefinement=gradient-based refinement (Algorithm 1)2020.12 | 92.92 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 96.6 | 90.43 | — | — | — | — | |
| SA-SSDmodel_source=pre-trained model provided by authors2020.12 | 92.84 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 96.56 | 90.36 | — | — | — | — | |
| CLOCS-PVCas2020.12 | 91.98 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 93.48 | 89.48 | — | — | — | — | |
| PV-RCNN2020.12 | 91.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 95.76 | 88.93 | — | — | — | — | |
| IPODrepresentation=BEV2018.12 | 86.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 88.3 | 84.6 | — | — | — | — | |
| VoxelNetrepresentation=BEV2018.12 | 84.81 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 89.6 | 78.57 | — | — | — | — | |
| F-PointNetrepresentation=BEV2018.12 | 84.02 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 88.16 | 76.44 | — | — | — | — | |
| MV3D (LiDAR)Modality=LiDAR2020.01 | 77.32 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 86.18 | 76.33 | — | — | — | — | |
| PL++: P-RCNN*Modality=Stereo, Extra Data=Scene Flow2020.01 | 64 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 82 | 57.3 | — | — | — | — | |
| DSGNModality=Stereo2020.01 | 63.91 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 83.24 | 57.83 | — | — | — | — | |
| PL++: AVOD*Modality=Stereo, Extra Data=Scene Flow2020.01 | 63.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 77 | 56 | — | — | — | — | |
| PL++: PIXOR*Modality=Stereo, Extra Data=Scene Flow2020.01 | 61.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 79.7 | 54.5 | — | — | — | — | |
| PL: AVOD*Modality=Stereo, Extra Data=Scene Flow2020.01 | 56.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 74.9 | 49 | — | — | — | — | |
| PL: F-PointNet*Modality=Stereo, Extra Data=Scene Flow2020.01 | 51.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 72.8 | 33.5 | — | — | — | — | |
| Stereo R-CNN*Modality=Stereo, Pre-training=ImageNet2020.01 | 48.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 68.5 | 41.5 | — | — | — | — | |
| MonoGRNetModality=Mono2020.01 | 28.39 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 43.75 | 23.87 | — | — | — | — | |
| AM3DModality=Mono2020.01 | 28.39 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 43.75 | 23.87 | — | — | — | — | |
| TriangulationModality=Stereo2020.01 | 21.88 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 29.22 | 18.83 | — | — | — | — | |
| M3D-RPNModality=Mono2020.01 | 21.18 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 25.94 | 17.9 | — | — | — | — | |
| MLFModality=Mono2020.01 | 19.54 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| 3DOPModality=Stereo2018.11 | 9.49 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 12.63 | 7.59 | — | — | — | — | |
| 3DOPModality=Stereo2020.01 | 9.49 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 12.63 | 7.59 | — | — | — | — | |
| OFT-NetModality=Mono2018.11 | 8.79 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 11.06 | 8.91 | — | — | — | — | |
| OFT-NetModality=Mono2020.01 | 8.79 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 11.06 | 8.91 | — | — | — | — | |
| Mono3DModality=Mono2018.11 | 5.19 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 5.22 | 4.13 | — | — | — | — | |
| 3DOPModality=Stereo2017.11 | — | 12.63 | 9.49 | 7.59 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| 3DOPInput=Image2018.02 | — | — | — | — | — | — | — | — | — | — | 55.04 | 41.25 | 34.55 | 12.63 | 9.49 | 7.59 | — | — | — | — | — | — | |
| 3DOPSensor=Stereo2019.02 | — | — | — | — | — | — | — | — | — | — | 55.04 | 41.25 | 34.55 | 12.63 | 9.49 | 7.59 | — | — | — | — | — | — | |
| Deep3DBoxInput=Image2018.02 | — | — | — | — | — | — | — | — | — | — | 29.96 | 24.91 | 19.46 | 9.01 | 7.94 | 6.57 | — | — | — | — | — | — | |
| Deep3DBoxSensor=Mono2019.02 | — | — | — | — | — | — | — | — | — | — | 30.02 | 23.77 | 18.83 | 9.99 | 7.71 | 5.3 | — | — | — | — | — | — | |
| HC-baselineModality=LiDAR2017.11 | — | 88.26 | 78.42 | 77.66 | 58.96 | 53.79 | 51.47 | 63.63 | 42.75 | 41.06 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Mono3DModality=Mono2017.11 | — | 5.22 | 5.19 | 4.13 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Mono3DInput=Image2018.02 | — | — | — | — | — | — | — | — | — | — | 30.5 | 22.39 | 19.16 | 5.22 | 5.19 | 4.13 | — | — | — | — | — | — | |
| Mono3DSensor=Mono2019.02 | — | — | — | — | — | — | — | — | — | — | 30.5 | 22.39 | 19.16 | 5.22 | 5.19 | 4.13 | — | — | — | — | — | — | |
| Multi-FusionSensor=Mono2019.02 | — | — | — | — | — | — | — | — | — | — | 55.02 | 36.73 | 31.27 | 22.03 | 13.63 | 11.6 | — | — | — | — | — | — | |
| Multi-FusionSensor=Stereo2019.02 | — | — | — | — | — | — | — | — | — | — | — | 53.56 | — | — | 19.54 | — | — | — | — | — | — | — | |
| MVModality=LiDAR, Input=BV+FV2017.11 | — | 86.18 | 77.32 | 76.33 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MVModality=LiDAR+Mono, Input=BV+FV+RGB2017.11 | — | 86.55 | 78.1 | 76.67 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MV3DInput=Fusion2018.02 | — | — | — | — | — | — | — | — | — | — | 96.34 | 89.39 | 88.67 | 86.55 | 78.1 | 76.67 | — | — | — | — | — | — | |
| Proposed Pipeline (MS-CNN)Backbone=MS-CNN, Input=Point Clouds2018.02 | — | — | — | — | — | — | — | — | — | — | 90.36 | 88.46 | 84.75 | 82.17 | 77.15 | 74.42 | — | — | — | — | — | — | |
| Proposed Pipeline (PC-CNN)Backbone=PC-CNN, Input=Point Clouds2018.02 | — | — | — | — | — | — | — | — | — | — | 88.31 | 83.74 | 79.62 | 83.61 | 77.36 | 69.61 | — | — | — | — | — | — | |
| Stereo R-CNNSensor=Stereo2019.02 | — | — | — | — | — | — | — | — | — | — | 87.13 | 74.11 | 58.93 | 68.5 | 48.3 | 41.47 | — | — | — | — | — | — | |
| VeloFCNModality=LiDAR2017.11 | — | 40.14 | 32.08 | 30.47 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| VeloFCNInput=LiDAR2018.02 | — | — | — | — | — | — | — | — | — | — | 79.68 | 63.82 | 62.8 | 40.14 | 32.08 | 30.47 | — | — | — | — | — | — | |
| VeloFCNSensor=LiDAR2019.02 | — | — | — | — | — | — | — | — | — | — | 79.68 | 63.82 | 62.8 | 40.14 | 32.08 | 30.47 | — | — | — | — | — | — | |
| VoxelNetModality=LiDAR2017.11 | — | 89.6 | 84.81 | 78.57 | 65.95 | 61.05 | 56.98 | 74.41 | 52.18 | 50.49 | — | — | — | — | — | — | — | — | — | — | — | — |