Academic Journals Database
Disseminating quality controlled scientific knowledge

Spin Image Based on Local Shape Similarity and Its Application

ADD TO MY LIST
 
Author(s): Yongwei Miao | Linghong Bao | Minyan Chen | Xudong Zhang

Journal: Transactions on Computer Science and Technology
ISSN 2327-090X

Volume: 2;
Issue: 3;
Start page: 40;
Date: 2013;
Original page

Keywords: Local Shape Similarity | Spin Image | Uniform Sampling | Visual Enhancement

ABSTRACT
Local shape similarity, an indicator to measure how similar a region of 3D shape is or dissimilar to another region, has been widely applied to computer graphics and computer vision. Different from traditional curvature map, a novel spin image based on local shape similarity measure is presented in this paper and its application on visual enhancement of 3D models is also given. To efficiently compare two different regions, the neighboring points for each surface vertex are firstly obtained by uniformly sampling along evenly distributed directions on the tangent plane. The spin images are constructed for these uniformly distributed sampling points and the local shape similarity measure can thus be calculated by comparing two spin images of different regions. Finally, due to our proposed local shape similarity definition, an efficient visual enhancement scheme is provided by incorporating our similarity measure into the color adjustment operation. Experimental results indicate that our spin image based on local shape similarity definition is robust and also contributes to visual enhancement.
Affiliate Program     

Tango Rapperswil
Tango Rapperswil