BEV Object Detection on Waymo Open Dataset (WOD) (val)
94.41Vehicle L1 APPillarNeXt-B
Evaluation results
| Method | Links | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| PillarNeXt-BFrames=32023.05 | 94.41 | 93.73 | 89.36 | 88.66 | 90.2 | 86.94 | 84.66 | 81.36 | 81.35 | 80.32 | 79.23 | 78.22 | — | — | — | |
| PillarNeXt-BFrames=12023.05 | 93.3 | 92.6 | 87.26 | 86.53 | 88.19 | 82.13 | 81.77 | 75.82 | 75.67 | 74.61 | 72.97 | 71.95 | — | — | — | |
| 3DAL+Frames=ALL, Protocol=Off-board method2023.05 | 93.3 | — | — | — | 86.32 | — | — | — | — | — | — | — | — | — | — | |
| SWFormerFrames=32023.05 | 92.6 | — | — | — | 87.5 | — | — | — | — | — | — | — | — | — | — | |
| PV-RCNN++*Frames=1, Model Stage=Two-stage2023.05 | 91.57 | — | — | — | 85.43 | — | — | — | 75.94 | — | — | — | — | — | — | |
| CPDLearning=Unsupervised2025.12 | — | — | — | — | — | — | — | — | — | — | — | — | 51.56 | 1,784 | 495 | |
| MODESTLearning=Unsupervised2025.12 | — | — | — | — | — | — | — | — | — | — | — | — | 13.87 | 10 | 30 | |
| OWLLearning=Unsupervised2025.12 | — | — | — | — | — | — | — | — | — | — | — | — | 57.76 | 3,085 | 3,372 | |
| OYSTERLearning=Unsupervised2025.12 | — | — | — | — | — | — | — | — | — | — | — | — | 23.67 | 20 | 10 |