Awesome

Backward Attentive Fusing Network with Local Aggregation Classifier for 3D Point Cloud Semantic Segmentation

This is the official implementation of BAF-LAC (TIP 2021), a novel point cloud semantic segmentation paradigm that introduces more context information. For technical details, please refer to:

Backward Attentive Fusing Network with Local Aggregation Classifier for 3D Point Cloud Semantic Segmentation<br /> Hui Shuai, Xiang Xu, Qingshan Liu. <br /> [paper] <br />

architecture

(1) Setup

This code has been tested with Python 3.6, Tensorflow 1.13.1, CUDA 10.0 on Ubuntu 16.04.

Clone the repository

git clone https://github.com/Xiangxu-0103/BAF-LAC.git && cd BAF-LAC

Setup python environment

conda create -n baflac python=3.6.8
conda activate baflac
pip install tensorflow-gpu==1.13.1
pip install -r helper_requirements.txt
sh compile_op.sh

(2) S3DIS

S3DIS dataset can be found <a href="https://docs.google.com/forms/d/e/1FAIpQLScDimvNMCGhy_rmBA2gHfDu3naktRm6A8BPwAWWDv-Uhm6Shw/viewform?c=0&w=1">here</a>. Download the files named "Stanford3dDataset_v1.2_Aligned_Version.zip". Uncompress the folder and move it to /home/data/S3DIS.

Preparing the dataset:

python utils/data_prepare_s3dis.py

Start 6-fold cross validation:

sh jobs_6_fold_cv_s3dis.sh

Move all the generated results (*.ply) in /test folder to /home/data/S3DIS/results, calculate the final mean IoU results:

python utils/6_fold_cv.py

(3) Semantic3D

7zip is required to uncompress the raw data in this dataset, to install p7zip:

sudo apt-get install p7zip-full

Download and extract the dataset. First, please specify the path of the dataset by changing the BASE_DIR in "download_semantic3d.sh"

sh utils/download_semantic3d.sh

Preparing the dataset:

python utils/data_prepare_semantic3d.py

Start training:

python main_Semantic3D.py --mode train --gpu 0

Evaluation:

python main_Semantic3D.py --mode test --gpu 0

Note:

Preferably with more than 64G RAM to process this dataset due to the large volume of point cloud

(4) SemanticKITTI

SemanticKITTI dataset can be found <a href="http://semantic-kitti.org/dataset.html#download">here</a>. Download the files related to semantic segmentation and extract everything into the same folder. Uncompress the folder and move it to /home/data/semantic_kitti/dataset.

Preparing the dataset

python utils/data_prepare_semantickitti.py

Start training:

python main_SemanticKITTI.py --mode train --gpu 0

Evaluation:

sh jobs_test_semantickitti.sh

Citation

If you find our work useful in your research, please consider citing:

@article{shuai2021backward,
    title={Backward Attentive Fusing Network With Local Aggregation Classifier for 3D Point Cloud Semantic Segmentation},
    author={Shuai, Hui and Xu, Xiang and Liu, Qingshan},
    journal={IEEE Transactions on Image Processing},
    volume={30},
    pages={4973--4984},
    year={2021},
    publisher={IEEE}
}

Acknowledge

Our code refers to <a href="https://github.com/QingyongHu/RandLA-Net">RandLA-Net</a>.