光学学报, 2017, 37 (12): 1215004, 网络出版: 2018-09-06   

基于贝叶斯推理的多尺度双目匹配方法 下载: 1013次

Multi-Scale Stereo Matching Based on Bayesian Reasoning
作者单位
上海交通大学机械与动力工程学院, 上海 200240
摘要
目前,虽然有许多双目匹配算法都可以实现较高的匹配精度,但其中能实现视频级实时计算的却极少。为此,提出一种基于贝叶斯推理的多尺度优化方法,在保证算法实时性的同时,可达到非常高的匹配精度。简单的局部匹配算法在设置不同窗口大小时,计算出来的视差图会反映不同尺度下场景的结构信息。基于此,提出一种对多张带有尺度信息和互补性的视差图进行基于贝叶斯推理的联合优化,从而得到高精度视差图的方法。对Middlebury立体视觉数据集的测试结果表明,本文算法在精度和效率上均优于其他实时算法。
Abstract
Most of the current stereo matching algorithms have high matching accuracy, but there are very few of them can realize real-time matching with video level frame rate. We present the multi-scale optimization algorithm based on Bayesian reasoning, which can be used to improve the matching accuracy, while maintaining the real-time performance. This algorithm obtains disparity maps with different scales by setting different window sizes. Based on this, the joint optimization based on Bayesian reasoning is proposed to optimize the disparity maps with scale information and complementarity. And then the high precision disparity maps are obtained. Test results of Middlebury stereo vision datasets show that the proposed algorithm has better accuracy and higher efficiency than several real-time algorithms.

曾灿灿, 任明俊, 肖高博, 殷跃红. 基于贝叶斯推理的多尺度双目匹配方法[J]. 光学学报, 2017, 37(12): 1215004. Cancan Zeng, Mingjun Ren, Gaobo Xiao, Yuehong Yin. Multi-Scale Stereo Matching Based on Bayesian Reasoning[J]. Acta Optica Sinica, 2017, 37(12): 1215004.

本文已被 2 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!