iBoW3D

Code for paper iBoW3D: Place Recognition Based on Incremental and General Bag of Words in 3D Scans (ICRA 2024).

Fig.1 The method overview

Video

• Bilibili
• Youtube

Prerequisites

We ran the code in Ubuntu 18.04 and we also used libraries listed below.

• Open3D
• OpenCV
• Eigen 3

Features

We use D3Feat to obtain keypoints and local features in advance.

Compile and Run

Before compile the package, some related paths should be added in the main.cpp file.

git clone https://github.com/NeSC-IV/iBoW3D.git
mkdir build
cd build
cmake ..
make
./ibow3d

Citation

@INPROCEEDINGS{10610036,
  author={Lin, Yuxiaotong and Chen, Jiming and Li, Liang},
  booktitle={2024 IEEE International Conference on Robotics and Automation (ICRA)}, 
  title={iBoW3D: Place Recognition Based on Incremental and General Bag of Words in 3D Scans}, 
  year={2024},
  pages={15981-15987},
  keywords={Measurement;Point cloud compression;Adaptation models;Solid modeling;Three-dimensional displays;Simultaneous localization and mapping;Databases},
  doi={10.1109/ICRA57147.2024.10610036}}

Acknowledgement

Thanks to the code of D3Feat.