一种基于贝叶斯的渐进式图像匹配框架研究

Progressive Image Matching Framework Based on Bayesian

提出了一种新的渐进式图像匹配框架,将图像匹配与图像概率更新结合在一起解决这一问题.该框架在当前匹配结果上,利用贝叶斯方法对最可信目标图像进行高效的重新估算,并且可保证提高后续的图像匹配效果.实验结果表明,与一次性图像匹配方法相比,所提方法的匹配性能更优.此外,即使是对外观发生变化及包含远离主体对象的图像场合,所提方法性能仍然健壮.

In this paper, we propose a novel progressive image matching framework which resolving the issue by combining probabilistic progression of graphs with matching of graphs. The framework efficiently re-estimates in a Bayesian manner the most plausible target graphs based on the current matching result, and guarantees to boost the matching objective at the subsequent graph matching. Experimental evaluation demonstrates that the performance of our approach is better than the oneshot graph matching. In addition, our method becomes robust to appearance variation as well as outliers, and generally applicable to generic objects with intra-class variation.