Compensated_Calibration_FPP

Published: 19 March 2021| Version 2 | DOI: 10.17632/66zwsnpcg3.2
Contributor:
Jian Wang

Description

The code and data show 3D reconstruction results of a checkerboard, using different calibration and reconstruction methods. (1) with general stereo calibration and linear reconstruction, (2) with general stereo calibration and approximately undistorted reconstruction, (3) with stereo calibration and undistorted reconstruction using nonlinear epipolar constraints, and (4) with the residual distortion-calibrated reconstruction. Instructions: Simply download all the files and data at any folder location of a computer. Change the working dictionary to the folder in MATLAB (2016 or after) and run the Demo.m to get the four reported reults. Please note that MATLAB Parallel Computing Toolbox is required for accellerated computation. The code has been tested with MATLAB 2016, under a Windows OS. To challenge the results using your own algorithm, please refer to the calibration images with the DOI: 10.21227/6f7p-jd52

Files

Steps to reproduce

Simply download all the files and data at any folder location of a computer. Change the working dictionary to the folder in MATLAB (2016 or after) and run the Demo.m to get the four reported reults. Please note that MATLAB Parallel Computing Toolbox is required for accellerated computation.

Institutions

Huazhong University of Science and Technology

Categories

Machine Vision, Structured Light, Instrument Calibration

Licence