Project Running Usage - HSqure/nutonomy_pointpillars GitHub Wiki

PointPillars Pytorch

Welcome to PointPillars(This is origin from nuTonomy/second.pytorch ReadMe.txt).

This repo demonstrates how to reproduce the results from PointPillars: Fast Encoders for Object Detection from Point Clouds (to be published at CVPR 2019) on the KITTI dataset by making the minimum required changes from the preexisting open source codebase SECOND.

This is not an official nuTonomy codebase, but it can be used to match the published PointPillars results.

WARNING: This code is not being actively maintained. This code can be used to reproduce the results in the first version of the paper, https://arxiv.org/abs/1812.05784v1. For an actively maintained repository that can also reproduce PointPillars results on nuScenes, we recommend using SECOND. We are not the owners of the repository, but we have worked with the author and endorse his code.

Getting Started!

This is a fork of SECOND for KITTI object detection and the relevant subset of the original README is reproduced here.

PART1.环境配置

1. Clone本工程

git clone https://github.com/HSqure/nutonomy_pointpillars.git

2.安装所需 Python 拓展包

推荐使用Anaconda package manager。

Anaconda:

conda create -n pointpillars python=3.6 anaconda
source activate pointpillars
conda install shapely pybind11 protobuf scikit-image numba pillow
conda install pytorch torchvision -c pytorch
conda install google-sparsehash -c bioconda

Pipe:

pip install --upgrade pip
pip install fire tensorboardX

Additionally, you may need to install Boost geometry:

sudo apt-get install libboost-all-dev

3. numba的CUDA设置

You need to add following environment variables for numba to ~/.bashrc:

export NUMBAPRO_CUDA_DRIVER=/usr/lib/x86_64-linux-gnu/libcuda.so
export NUMBAPRO_NVVM=/usr/local/cuda/nvvm/lib64/libnvvm.so
export NUMBAPRO_LIBDEVICE=/usr/local/cuda/nvvm/libdevice

4. PYTHON PATH

Add nutonomy_pointpillars/ to your PYTHONPATH.

export PYTHONPATH=$PYTHONPATH:/your_root_path/nutonomy_pointpillars/

PART2.KITTI数据集准备

KITTI Dataset概览:

$KITTI_DATASET_ROOT
 ├── testing # 7580 test data
 │   ├── calib
 │   ├── image_2  # for visualization
 │   ├── velodyne # point cloud bin file
 │   └── velodyne_reduced # empty directory
 └── training
     ├── calib  # 7481 train data
     ├── image_2
     ├── label_2
     ├── velodyne
     └── velodyne_reduced # empty directory

其中training目录是有标注的数据（标注数据目录是label_2），工程中training和evaluate用的都是training目录下的数据；testing目录下面都是没有标注的数据，可以用来测测模型的检测效果（需要自己写可视化代码看看最终的检测结果）。

在工程根目录下进入子目录second，通过下面几条命令创建info数据。

1.创建.pkl参数文件

python create_data.py create_kitti_info_file --data_path=$KITTI_DATASET_ROOT

在$KITTI_DATASET_ROOT目录下创建: kitti_infos_train.pkl、 kitti_infos_val.pkl、 kitti_infos_trainval.pkl、 kitti_infos_test.pkl 四个.bin文件，每个.bin文件包含了图片的路径、calib中Camera和Lidar的标定参数等。

2.生成裁减后的数据副本

python create_data.py create_reduced_point_cloud --data_path=$KITTI_DATASET_ROOT

在$KITTI_DATASET_ROOT/training/velodyne_reduced和$KITTI_DATASET_ROOT/testing/velodyne_reduced目录下分别创建它们同级velodyne目录中点云.bin文件的reduce版本，去掉了点云数据中一些冗余的背景等数据，可以认为是经过裁剪的点云数据。

3.生成GroundTruth数据

python create_data.py create_groundtruth_database --data_path=$KITTI_DATASET_ROOT

在$KITTI_DATASET_ROOT目录下创建kitti_dbinfos_train.pkl数据。在训练代码中的input_cfg.database_sampler变量中用到。

完成步骤后的dataset概览：

$KITTI_DATASET_ROOT
 ├── kitti_dbinfos_train.pkl  # step 03
 ├── kitti_infos_test.pkl # step 01
 ├── kitti_infos_train.pkl # step 01
 ├── kitti_infos_trainval.pkl # step 01
 ├── kitti_infos_val.pkl # step 01
 ├── gt_database # step 03
 ├── testing
 │   ├── calib
 │   ├── image_2  
 │   ├── velodyne 
 │   └── velodyne_reduced # step 02
 └── training
     ├── calib 
     ├── image_2
     ├── label_2
     ├── velodyne
     └── velodyne_reduced # step 02

4. 修改配置文件

配置文件路径:second/configs/pointpillars/car/xyres_16.proto。 需在以下部份包含前面生成的所有的数据集.pkl文件。

train_input_reader: {
  ...
  database_sampler {
    database_info_path: "/path/to/kitti_dbinfos_train.pkl"
    ...
  }
  kitti_info_path: "/path/to/kitti_infos_train.pkl"
  kitti_root_path: "KITTI_DATASET_ROOT"
}
...
eval_input_reader: {
  ...
  kitti_info_path: "/path/to/kitti_infos_val.pkl"
  kitti_root_path: "KITTI_DATASET_ROOT"
}

PART3.模型训练与评估

1. 训练 Training

cd second
python ./pytorch/train.py train --config_path=./configs/pointpillars/car/xyres_16.proto --model_dir=/path/to/model_dir
  • If you want to train a new model, make sure "/path/to/model_dir" doesn't exist.
  • If "/path/to/model_dir" does exist, training will be resumed from the last checkpoint.
  • Training only supports a single GPU.
  • Training uses a batchsize=2 which should fit in memory on most standard GPUs.
  • On a single 1080Ti, training xyres_16 requires approximately 20 hours for 160 epochs.

2. 评估 Evaluate

cd second/
python pytorch/train.py evaluate --config_path= configs/pointpillars/car/xyres_16.proto --model_dir=/path/to/model_dir
  • Detection result will saved in model_dir/eval_results/step_xxx.
  • By default, results are stored as a result.pkl file. To save as official KITTI label format use --pickle_result=False.

参考