SHREC 2022: Fitting and recognition of simple geometric primitives on point clouds
This repository contains the routines that were used to produced the results in [1], which were tested on a 2019 MacBook pro running MacOS Monterey version 12.4. The numbering of figures and tables refers to the ArXiv version of the paper. The dataset, the ground truth and the results submitted by the participants of the track can be found at:
https://github.com/chiararomanengo/SHREC2022.git
Before running
The folders foung in the link above should be placed as follows:
- The folder
datasetshould be placed at the same level asresults. - The folder
methodsshould be placed insideresults. Note that these folders are collected in a separate GitHub repository as the size limit for GRSI is set at 500 MB.
The files
Figures 8-10 can be generated by running the python scripts "figure8.py", "figure9.py", "figure10a.py", "figure10b.py" and "figure10c.py". No manual change is required.
Classification measures in Table 1 and in Tables 3-12 are obtained by running the python script "classification_measures.py". Table 1 is obtained by setting config['specify']=0 inside the file. Similarly, Tables 3-12 are obtained by setting config['specify']=k, with k=1,...,10.
Fitting and recognition measures in Table 2 and in Tables 13-22 are obtained by running the python scripts "fitting_measures_Haus_analysis.py", "fitting_measures_L2.py" and "fitting_measures_MFE.py". Table 2 is obtained by setting config['specify']=0 inside the files. Similarly, Tables 13-22 are obtained by setting config['specify']=k, with k=1,...,10.
References
[1] Chiara Romanengo, Andrea Raffo, Silvia Biasotti, Bianca Falcidieno, Vlassis Fotis, Ioannis Romanelis, Eleftheria Psatha, Konstantinos Moustakas, Ivan Sipiran, Quang-Thuc Nguyen, Chi-Bien Chu, Khoi-Nguyen Nguyen-Ngoc, Dinh-Khoi Vo, Tuan-An To, Nham-Tan Nguyen, Nhat-Quynh Le-Pham, Hai-Dang Nguyen, Minh-Triet Tran, Yifan Qie, Nabil Anwer, "SHREC 2022: Fitting and recognition of simple geometric primitives on point clouds", Computers & Graphics, Volume 107, 2022, Pages 32-49, https://doi.org/10.1016/j.cag.2022.07.004.