Title:	EVOLUTIONARY GENERATION OF 3D SHAPE MODELS
Author(s):	Satoshi Yonemoto and Masaya Sugawa
ISBN:	978-989-8533-66-1
Editors:	Yingcai Xiao and Ajith P. Abraham
Year:	2017
Edition:	Single
Keywords:	Evolutionary computation, 3D shape modeling, part crossover and genetic algorithm
Type:	Full Paper
First Page:	47
Last Page:	54
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	We introduce a method for generating evolutionary 3D shape models. Given several initial models and the initial model's content ratio, our method generates novel and plausible 3D shapes. Our method is based on evolutionary computation (genetic algorithm). Our model consists of several parts defined by a structure graph. In the evolutionary operations, two-step crossover and mutation are executed. In part crossover, type of parts may be exchanged. Properties such as scale or detailed shape are usually inherited from the parents. In mutation, a new type of part which does not exist in the initial models can be appeared. We demonstrate the effectiveness of our evolutionary generation of 3D shapes, which reflects a user's intention and takes diversity of 3D shapes into account simultaneously.
	Go Back