2012 International Conference on Computational Intelligence and Software Engineering (CiSE 2012)(E-BOOK)

Wuhan,China,2012-12-142012-12-162012

ISBN: 978-1-61896-036-8 Scientific Research Publishing

E-Book 275pp Pub. Date: December 2012

Category: Computer Science & Communications

Price: $100

Title: Fast Template Matching Method for 3D Object Detection with RGBD Data
Source: 2012 International Conference on Computational Intelligence and Software Engineering (CiSE 2012)(E-BOOK) (pp 72-75)
Author(s): Hailong Cao, Shanghai Jiaotong University, China
Xubo Yang, Shanghai Jiaotong University, China
Abstract: In this paper, we present an improved template matching algorithm for fast 3D object detection with RGBD data. Our detection algorithm is based on a template matching algorithm known as “Dominant Orientation Templates” (DOT). We extend it on depth image and at the same time we define a few key grids by using the information of gradient orientations of the RGB and depth templates. These key grids are employed to quickly filter most negative matching areas thus efficiently reduce the matching times without loss of accuracy. We test our work both on RGB-D Object Dataset and real scenes, both of them return good results.
Free SCIRP Newsletters
Copyright © 2006-2024 Scientific Research Publishing Inc. All Rights Reserved.
Top