BEV-MAE: Bird's Eye View Masked Autoencoders for Outdoor Point Cloud Pre-training
This is the official implementation of BEV-MAE. The training code is coming soon.
Model
We release the pre-training weights of VoxelNet on Waymo dataset.
| pre-trained 3D backbone | Dataset | Weights |
|---|---|---|
| VoxelNet | Waymo (20% data) | Google_drive |
| VoxelNet | Waymo (full data) | Google_drive |
Our code is base on OpenPCDet. To use our pre-trained weights, please refer to INSTALL.md for installation and follow the instructions in GETTING_STARTED.md to train the model.
Contact Us
If you have any problem about this work, please feel free to reach us out at zwlin@pku.edu.cn.