Cross-trees, edge and superpixel priors-based cost aggregation for stereo matching
作者：
Highlights：
• We propose a novel cross-trees structure to perform the non-local cost aggregation.
• The trees׳ constructions are unique and independent on the image itself.
• We propose and incorporate two priors into the non-local framework.
• New edge weights function is designed according to the trees and the priors.
• Optimal support regions can be chose by cutting the cost aggregation flow on paths.
摘要
Highlights•We propose a novel cross-trees structure to perform the non-local cost aggregation.•The trees׳ constructions are unique and independent on the image itself.•We propose and incorporate two priors into the non-local framework.•New edge weights function is designed according to the trees and the priors.•Optimal support regions can be chose by cutting the cost aggregation flow on paths.
论文关键词：Stereo matching,Cost aggregation,Image filtering,Spanning trees
论文评审过程：Received 9 April 2014, Revised 6 November 2014, Accepted 5 January 2015, Available online 14 January 2015.
论文官网地址：https://doi.org/10.1016/j.patcog.2015.01.002
