Awesome
LiDAR R-CNN: An Efficient and Universal 3D Object Detector
Introduction
This is the official code of LiDAR R-CNN: An Efficient and Universal 3D Object Detector. In this work, we present LiDAR R-CNN, a second stage detector that can generally improve any existing 3D detector. We find a common problem in Point-based RCNN, which is the learned features ignore the size of proposals, and propose several methods to remedy it. Evaluated on WOD benchmarks, our method significantly outperforms previous state-of-the-art.
中文介绍:https://zhuanlan.zhihu.com/p/359800738
News
- We provide the training code for multi-frame setting, and show 3 frame results based PointPillars.
Requirements
All the codes are tested in the following environment:
- Linux (tested on Ubuntu 16.04)
- Python 3.6+
- PyTorch 1.5 or higher (tested on PyTorch 1.5, 6, 7)
- CUDA 10.1
To install pybind11:
git clone git@github.com:pybind/pybind11.git
cd pybind11
mkdir build && cd build
cmake .. && make -j
sudo make install
To install requirements:
pip install -r requirements.txt
apt-get install ninja-build libeigen3-dev
Install LiDAR_RCNN
library:
python setup.py develop --user
Cuda Extensions:
# Rotated IOU
cd src/LiDAR_RCNN/ops/iou3d/
python setup.py build_ext --inplace
Preparing Data
Please refer to data processer to generate the proposal data.
Training
After preparing WOD data, we can train the vehicle only model in the paper, run this command:
python -m torch.distributed.launch --nproc_per_node=4 tools/train.py --cfg config/lidar_rcnn.yaml --name lidar_rcnn
For 3 class in WOD:
python -m torch.distributed.launch --nproc_per_node=8 tools/train.py --cfg config/lidar_rcnn_all_cls.yaml --name lidar_rcnn_all
The models and logs will be saved to work_dirs/outputs
.
NOTE: for multi-frame training, please set MODEL.Frame = n
in config.
Evaluation
To evaluate, run distributed testing with 4 gpus:
python -m torch.distributed.launch --nproc_per_node=4 tools/test.py --cfg config/lidar_rcnn.yaml --checkpoint outputs/lidar_rcnn/checkpoint_lidar_rcnn_59.pth.tar
python tools/create_results.py --cfg config/lidar_rcnn.yaml
Note that, you should keep the nGPUS
in config equal to nproc_per_node
.This will generate a val.bin
file in the work_dir/results
. You can create submission to Waymo server using waymo-open-dataset code by following the instructions here.
Results
Our model achieves the following performance on:
Waymo Open Dataset Challenges (3D Detection)
Proposals from | Class | Frame/Channel | 3D AP L1 Vehicle | 3D AP L1 Pedestrian | 3D AP L1 Cyclist |
---|---|---|---|---|---|
PointPillars | Vehicle | 1 / 1x | 75.6 | - | - |
PointPillars | Vehicle | 1 / 2x | 75.6 | - | - |
PointPillars | Vehicle | 3 / 2x | 77.8 | - | - |
SST | Vehicle | 3 / 2x | 78.6 | - | - |
PointPillars | 3 Class | 1 / 1x | 73.4 | 70.7 | 67.4 |
PointPillars | 3 Class | 1 / 2x | 73.8 | 71.9 | 69.4 |
Proposals from | Class | Frame/Channel | 3D AP L2 Vehicle | 3D AP L2 Pedestrian | 3D AP L2 Cyclist |
---|---|---|---|---|---|
PointPillars | Vehicle | 1 / 1x | 66.8 | - | - |
PointPillars | Vehicle | 1 / 2x | 67.9 | - | - |
PointPillars | Vehicle | 3 / 2x | 69.1 | - | - |
SST | Vehicle | 3 / 2x | 69.9 | - | - |
PointPillars | 3 Class | 1 / 1x | 64.8 | 62.4 | 64.8 |
PointPillars | 3 Class | 1 / 2x | 65.1 | 63.5 | 66.8 |
Note: The proposals provided by PointPillars are detected on 1 frame points cloud.
Citation
If you find our paper or repository useful, please consider citing
@article{li2021lidar,
title={LiDAR R-CNN: An Efficient and Universal 3D Object Detector},
author={Li, Zhichao and Wang, Feng and Wang, Naiyan},
journal={CVPR},
year={2021},
}
Acknowledgement
This project draws on the following codebases.