Building facade detection, segmentation, and parameter estimation for mobile robot stereo vision
作者：
Highlights：
• We perform automatic modeling of building facades from single-view stereo.
• Goal is to provide semantic landmarks for mobile platform localization.
• Use of novel appearance model with stereo features and plane fitting to disparity.
• Facade segmentation by graphical model: hierarchical Markov Random Field.
• Good performance in segmenting and modeling major facades in the scene.
摘要
•We perform automatic modeling of building facades from single-view stereo.•Goal is to provide semantic landmarks for mobile platform localization.•Use of novel appearance model with stereo features and plane fitting to disparity.•Facade segmentation by graphical model: hierarchical Markov Random Field.•Good performance in segmenting and modeling major facades in the scene.
论文关键词：Stereo vision,Mobile robot perception,Hierarchical Markov random field,Building facade detection,Model-based stereo vision
论文评审过程：Received 11 January 2012, Revised 22 October 2012, Accepted 29 August 2013, Available online 5 September 2013.
论文官网地址：https://doi.org/10.1016/j.imavis.2013.08.006
