Regional image correspondence matching method for SIMD processing

Author(s):  
Olli Lahdenoja ◽  
Mika Laiho
2015 ◽  
Vol 1 (1) ◽  
Author(s):  
Francis Li ◽  
Alexander Wong ◽  
John Zelek

<p>This work implements a method to improve correspondence matching<br />in stereo vision by using varying illumination intensities from an<br />external light source. By iteratively increasing the light intensity on<br />the scene, different parts of the scene become saturated in the left<br />and right images. These saturated areas are assumed to correspond<br />to each other, greatly reducing the search space for stereo<br />correspondence and increasing robustness to erroneous matches.<br />The stereo camera and light source used in this work is the DUO3D<br />camera by Code Laboratories. Visually, experimental results show<br />the resultant point clouds from the proposed method is less noisy<br />with fewer outliers compared to standard block matching method,<br />but produces fewer matches.</p>


1985 ◽  
Vol 46 (C4) ◽  
pp. C4-321-C4-329 ◽  
Author(s):  
E. Molinari ◽  
G. B. Bachelet ◽  
M. Altarelli

2007 ◽  
Vol 127 (4) ◽  
pp. 573-580 ◽  
Author(s):  
Toshiyuki Sawa ◽  
Yuji Nakata ◽  
Mitsuo Tsurugai ◽  
Shigenari Sugiyama

2018 ◽  
Vol 2018 (1) ◽  
pp. 313-317
Author(s):  
Yuechen Zhu ◽  
Qiyan Zhal ◽  
Ming Ronnier Luo

2013 ◽  
Vol 26 (6) ◽  
pp. 521-528 ◽  
Author(s):  
Wenjun Xiong ◽  
Kejun Xu ◽  
Min Fang ◽  
Qili Hou ◽  
Cui Liu

Sign in / Sign up

Export Citation Format

Share Document