Shape description of two superquadrics from range data using GA
Hideki Tanahashi, Kazuhiko Yamamoto, Kunihito Kato · IEEJ Transactions on Industry Applications · 1999
In this paper, we propose a method for recovering shape descriptions of three-dimensional object from range image by using genetic algorithm (GA). Two superquadrics models have been employed for description of a 3D shape. The model parameters, which defined the shape, are obtained at the same time without using the division method. The fitting energy evaluated the difference in both depth from the object center and the distribution of the normal vector between input range image and the model surfaces. As a result, the experiment demonstrates that the three-dimensional object was reconstructed from both artificial and measured range data.