Paper Info
Title
Efficient contrast invariant stereo correspondence using dynamic programming with vertical constraint
Abstract
In this paper, we propose a dense stereo algorithm based on the census transform and improved dynamic programming (DP). Traditional scanline-based DP algorithms are the most efficient ones among global algorithms, but are well-known to be affected by the streak effect. To solve this problem, we improve the traditional three-state DP algorithm by taking advantage of an extended version of sequential vertical consistency constraint. Using this method, we increase the accuracy of the disparity map greatly. Optimizations have been made so that the computational cost is only increased by about 20%, and the additional memory needed for the improvement is negligible. Experimental results show that our algorithm outperforms many state-of-the-art algorithms with similar efficiency on Middlebury College’s stereo Web site. Besides, the algorithm is robust enough for image pairs with utterly different contrasts by using of census transform as the basic match metric.
Year
DOI
Venue
2008
10.1007/s00371-007-0177-9
The Visual Computer
Keywords
Field
DocType
middlebury college,global algorithm,stereo correspondence · dynamic programming · vertical constraint · computer vision,traditional scanline-based dp algorithm,dense stereo algorithm,state-of-the-art algorithm,traditional three-state dp algorithm,dynamic programming,additional memory,stereo web site,vertical constraint,computational cost,efficient contrast invariant stereo,basic match metric,computer vision
Computer vision,Dynamic programming,Mathematical optimization,Computer science,Streak,Contrast (statistics),Census transform,Invariant (mathematics),Artificial intelligence,Web site,Scan line
Journal
Volume
Issue
ISSN
24
1
1432-2315
Citations 
PageRank 
References 
4
0.39
21
Authors
4
Name
Order
Citations
PageRank
Zhiliang Xu114316.90
Lizhuang Ma2498100.70
Masatoshi Kimachi372.59
Masaki Suwa4599.06