Home

This web site presents a Computer Science Honours Project, entitled "Biologically Inspired Plant Modelling for Synthetic Imagery". The project was undertaken by Aidan Lane under the supervision of Dr. Jon McCormack at the School of Computer Science and Software Engineering (CSSE), located at Monash University, Clayton campus, Australia.

Please download the thesis for the full details regarding the project.
If you have any questions, please contact Aidan Lane via e-mail: alane@csse.monash.edu.au.

Thesis Abstract

The desire to create models of plants has increasingly become a cross-disciplinary process, shared by artists, designers, biologists and computer scientists. While various methods exist to assist with this modelling, they have been overly complex to use or inaccurate in their results. Current methods include L-systems and applications of evolution-based computation and visual graph-based languages. Thus, the aim of this study is to develop a new method that focuses on the way that branching structures would be observed in nature. In addition, a new method must support visualisation of the resulting model, in order to produce the synthetic imagery.
This thesis presents a new, biologically inspired method for modelling plants. It allows plant architectures to be described concisely, using existing concepts, such as branching point types and a graph that is used to specify the topology. This form of model enables realistic branching structures to be modelled and should appeal to a broader audience of modellers, in comparison to previous systems. Furthermore, these branching structures can include stochastic variation of individual components, which enables entire forests of plants to be created from a single model. The implementation allows the models to be exported into a geometric form, suitable for use in professional rendering packages.