Student research opportunities
Image based 3D plant modeling
Project Code: CECS_132
This project is available at the following levels:
Honours, Summer Scholar, Masters, PhD
Keywords:
Computer vision, 3D reconstruction, Multi-view Geometry, Structure from motion, computer graphics.
Supervisor:
Dr Hongdong LiOutline:
This project is aimed at developing a basic 3D computer vision system that is able to reconstruct a 3D model of a plant (flower/herb/trees) from its multiple images.
Two key technique components will be investigated in this research project. (1) structure from motion, which will be used to compute 3D coordinates of the feature points on the plant; (2) image segmentation, which will be used to extract
meaningful part of the plant (such as leaves, branches). Since we need to model the individual leaves, these leaves need to be segmented out from the images.
Goals of this project
Develop a Matlab based 3D vision reconstruction software system which is able to reconstruct 3D point clouds from several images of a given plant (e.g. flower,herb,crop) being imaged.
Develop new computer vision algorithms suitable for plant modeling.
The outcomes of this project will be helpful for plant structural phenomics research.
Requirements/Prerequisites
Good knowledge of linear algebra, matrix computation, and elementary geometry.
Basic knowledge of digital image processing.
Good knowledge of Matlab or C/C++.
Student Gain
Good for students who are interested in 3D computer graphics, computer vision, image processing, pattern recognition, and 3D animation.
Research experiences in leading research university and NICTA.
Background Literature
Main references:
[1] Image-based tree modeling, SIGGRAPH 2007
[2] Image based plant modeling, SIGGRAPH 2006
[3] Any computer vision text-book.
