Student research opportunities
Light field camera for robots
Project Code: CECS_1173
This project is available at the following levels:
CS single semester, Engn4200, Engn R&D, Honours, Masters
Keywords:
Computational photography, plenoptic camera
Supervisors:
Assoc Professor Hongdong LiDr Chuong Nguyen
Outline:
Light field cameras is gaining attention thanks to its novel capabilities including refocusing image and occlusion removal. Commercial light field cameras are expensive and difficult to customise for specific applications. There is a need for affordable and customisable camera system for specific applications.
Goals of this project
This project will create a custom field camera array with on-board computers. The camera array can be easily customised for different specialised applications. One application is detection of fruits behind leaves for a robot arm to pick. Other applications include medical and defence.
Requirements/Prerequisites
Computer Vision (ENGN4528)
Experience with matrix algebra, MATLAB, Python or C/C++.
Student Gain
Student will gain experience with light field imaging, camera hardware, camera calibration and parallel computing.
Background Literature
Lytro - ILLUM https://www.lytro.com/illum/
Xu, Yichao, et al. "Mobile Camera Array Calibration for Light Field Acquisition." arXiv preprint arXiv:1407.4206 (2014). http://arxiv.org/pdf/1407.4206
Dansereau, Donald G., Oscar Pizarro, and Stefan B. Williams. "Decoding, calibration and rectification for lenselet-based plenoptic cameras." Computer Vision and Pattern Recognition (CVPR), 2013 IEEE Conference on. IEEE, 2013. http://www.cv-foundation.org/openaccess/content_cvpr_2013/papers/Dansereau_Decoding_Calibration_and_2013_CVPR_paper.pdf
