alalemp/bkisemanticmapping

Bayesian Spatial Kernel Smoothing for Scalable Dense Semantic Mapping

BKISemanticMapping

Bayesian Spatial Kernel Smoothing for Scalable Dense Semantic Mapping

Getting Started

Building with catkin

catkin_ws/src$ git clone https://github.com/ganlumomo/BKISemanticMapping
catkin_ws/src$ cd ..
catkin_ws$ catkin_make
catkin_ws$ source ~/catkin_ws/devel/setup.bash

Building using Intel C++ compiler (optional for better speed performance)

catkin_ws$ source /opt/intel/compilers_and_libraries/linux/bin/compilervars.sh intel64
catkin_ws$ catkin_make -DCMAKE_C_COMPILER=icc -DCMAKE_CXX_COMPILER=icpc
catkin_ws$ source ~/catkin_ws/devel/setup.bash

Running the Demo

$ roslaunch semantic_bki toy_example_node.launch

Semantic Mapping using KITTI dataset

Download Data

Please download data_kitti_15 and uncompress it into the data folder.

Running

$ roslaunch semantic_bki kitti_node.launch

You will see semantic map in RViz. It also projects 3D grid onto 2D image for evaluation, stored at data/data_kitti_05/reproj_img.

Evaluation

Evaluation code is provided in kitti_evaluation.ipynb. You may modify the directory names to run it.

Semantic Mapping using SemanticKITTI dataset

Download Data

Please download semantickitti_04 and uncompress it into the data folder.

Running

$ roslaunch semantic_bki semantickitti_node.launch

You will see semantic map in RViz. It also query each ground truth point for evaluation, stored at data/semantickitti_04/evaluations.

Evaluation

Evaluation code is provided in semantickitti_evaluation.ipynb. You may modify the directory names to run it, or follow the guideline in semantic-kitti-api for evaluation.

Relevant Publications

If you found this code useful, please cite the following:

Bayesian Spatial Kernel Smoothing for Scalable Dense Semantic Mapping (PDF)

@article{gan2019bayesian,
  title={Bayesian Spatial Kernel Smoothing for Scalable Dense Semantic Mapping},
  author={Gan, Lu and Zhang, Ray and Grizzle, Jessy W and Eustice, Ryan M and Ghaffari, Maani},
  journal={arXiv preprint arXiv:1909.04631},
  year={2019}
}

Learning-Aided 3-D Occupancy Mapping with Bayesian Generalized Kernel Inference (PDF)

@article{Doherty2019,
  doi = {10.1109/tro.2019.2912487},
  url = {https://doi.org/10.1109/tro.2019.2912487},
  year = {2019},
  publisher = {Institute of Electrical and Electronics Engineers ({IEEE})},
  pages = {1--14},
  author = {Kevin Doherty and Tixiao Shan and Jinkun Wang and Brendan Englot},
  title = {Learning-Aided 3-D Occupancy Mapping With Bayesian Generalized Kernel Inference},
  journal = {{IEEE} Transactions on Robotics}
}