jixingwu / Edit_Distance

the core programe of topology aware object-level semantic mapping towards more robust loop closure for comparison

Edit_Distance

The repository is used to release unofficial code for "Topolog aware object-level semantic mapping towards more robust loop closure" for comparison. The programs of visible object detection and initial pose estimation are VisualDet3D and VINS-Fusion respectively.

Setup

1.1 Ubuntu and ROS

Ubuntu 64-bit 18.04. ROS Melodic.

1.2 VisualDet3D

• environment setup

pip3 install -r requirement.txt

• config and path setup

Please modify the path and parameters in visualDet3D/config.py* files.

• stereo image as input training

./launchers/det_precompute.sh config/$CONFIG_FILE.py train
./launcher/train.sh  config/$CONFIG_FILE.py 0 $experiment_name

• stereo image as input testing

./launchers/det_precompute.sh config/$CONFIG_FILE.py test
./launchers/eval.sh config/$CONFIG_FILE.py 0 visualDet3D/workdirs/Stereo3D/checkpoint/Stereo3D_latest.pth test

• the evaluate results

the object detected results files should be stored in sequences/xx. VINS-Fusion can read the results.

1.3 VINS-Fusion

• environment setup

ceres solver: follow Ceres Installation

• KITTI Example

 roslaunch vins vins_rviz.launch
  rosrun vins kitti_odom_test ~/catkin_ws/src/VINS-Fusion/config/kitti_odom/kitti_config00-02.yaml YOUR_DATASET_FOLDER/sequences/00/

1.4 Edit_distance

With the object detection results and pose estimation results, you can test your own loop detection and store its results as two files, such as graph1.txt, graph2.txt. To compute the graphs edit distance, you can run:

./ED path_to_graph1/graph1.txt path_to_graph2/graph2.txt

The executable program will output the minimum edit distance between graph1 and graph2.

Related Paper:

if you think the repository and codes are useful, please cite the following papers.

@ARTICLE{9327478,
  author={Y. {Liu} and Y. {Yuan} and M. {Liu}},
  journal={IEEE Robotics and Automation Letters}, 
  title={Ground-aware Monocular 3D Object Detection for Autonomous Driving}, 
  year={2021},
  doi={10.1109/LRA.2021.3052442}}
  
@article{qin2019fusion,
  author={Tong Qin and Jie Pan and Shaozu Cao and Shaojie Shen},
  title={A General Optimization-based Framework for Local Odometry Estimation with Multiple Sensors},
  journal={ArXiv},
  year={2019},
  volume={abs/1901.03638}}
  
@ARTICLE{lin_topology_2021,  
  author={Lin, Shiqi and Wang, Jikai and Xu, Meng and Zhao, Hao and Chen, Zonghai},  
  journal={IEEE Robotics and Automation Letters},   
  title={Topology Aware Object-Level Semantic Mapping Towards More Robust Loop Closure},   
  year={2021}
  doi={10.1109/LRA.2021.3097242}}