DocumentCode :
3518623
Title :
Binary confidence evaluation for a stereo vision based depth field processor SoC
Author :
Motten, Andy ; Claesen, Luc ; Pan, Yun
Author_Institution :
Expertise Centre for Digital Media, Hasselt Univ. - tUL - IBBT, Diepenbeek, Belgium
fYear :
2011
fDate :
28-28 Nov. 2011
Firstpage :
456
Lastpage :
460
Abstract :
This paper presents a methodology to construct a binary confidence value for every pixel of a depth map. We start by constructing 72 different confidence metrics, including the traditional ones and new metrics based on neighborhood information. Construction of the binary confidence value from these metrics is hence viewed as a two-class classification problem where we evaluated three different classifiers, with increasing complexity. Only metrics and classifiers that are suitable for VLSI hardware implementation will be evaluated. Evaluation of the constructed classifiers is performed on an indoor dataset of Stereo Images.
Keywords :
VLSI; computer vision; image classification; image matching; stereo image processing; system-on-chip; VLSI hardware implementation; binary confidence evaluation; confidence metrics; depth field processor SoC; depth map; neighborhood information; stereo image; stereo vision; two-class classification problem; Artificial neural networks; Computer vision; Hardware; Image color analysis; Measurement; Stereo vision; System-on-a-chip; SoC; binary adaptable window; computer vision; stereo confidence; stereo matching;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Pattern Recognition (ACPR), 2011 First Asian Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4577-0122-1
Type :
conf
DOI :
10.1109/ACPR.2011.6166593
Filename :
6166593
Link To Document :
بازگشت