Det3D

Det3D is the first 3D Object Detection toolbox which provides off the box implementations of many 3D object detection algorithms such as PointPillars, SECOND, PIXOR, etc, as well as state-of-the-art methods on major benchmarks like KITTI(ViP) and nuScenes(CBGS). Key features of Det3D include the following aspects:

  • Multi Datasets Support: KITTI, nuScenes, Lyft
  • Point-based and Voxel-based model zoo
  • State-of-the-art performance
  • DDP & SyncBN

Model Zoo and Baselines

Second on KITTI(val) Dataset

car  AP @0.70, 0.70,  0.70:
bbox AP:90.54, 89.35, 88.43
bev  AP:89.89, 87.75, 86.81
3d   AP:87.96, 78.28, 76.99
aos  AP:90.34, 88.81, 87.66

PointPillars on KITTI(val) Dataset

car  [email protected],  0.70,  0.70:
bbox AP:90.63, 88.86, 87.35
bev  AP:89.75, 86.15, 83.00
3d   AP:85.75, 75.68, 68.93
aos  AP:90.48, 88.36, 86.58

To Be Released

  1. PointPillars on NuScenes(val) Dataset
  2. CGBS on NuScenes(val) Dataset
  3. CGBS on Lyft(val) Dataset

Currently Support

  • Models
    • [x] VoxelNet
    • [x] SECOND
    • [x] PointPillars
  • Features
    • [x] Multi task learning & Multi-task Learning
    • [x] Distributed Training and Validation
    • [x] SyncBN
    • [x] Flexible anchor dimensions
    • [x] TensorboardX
    • [x] Checkpointer & Breakpoint continue
    • [x] Self-contained visualization
    • [x] Finetune
    • [x] Multiscale Training & Validation
    • [x] Rotated RoI Align

TODO List

  • Models
    • [ ] PointRCNN
    • [ ] PIXOR

GitHub