一种基于分层学习的关键点匹配算法

A Keypoint Matching Method Based on Hierarchical Learning

关键点匹配技术是计算机视觉中的一项重要技术,其最主要的问题是寻找一种快速鲁棒的关键点匹配算法。该文提出了一种基于分层学习的二值描述符匹配算法。该方法将二值描述符学习过程分为粗细两个层次,结合了固定点抽样模式和随机抽样模式的优点,提高了学习效率;另外,该方法建立了更加合理的点对辨识模型并将其应用到关键点匹配算法中,提高了匹配精度。实验结果表明,在低计算复杂度下,该方法的匹配精度仍优于其它经典的二值描述符匹配算法。

Keypoint matching is an important task of computer vision and the major problem is to find a fast and robust keypoints algorithm. This paper presents a binary descriptor matching algorithm based on hierarchical learning method. The descriptor learning process is divided into two levels of coarse and fine, which combines the advantages of the fixed-point sampling mode and random sampling mode, and the process enhances the performance of learning. Meanwhile, a more reasonable point-pair identification model is built and applied into the keypoint matching algorithm which improves the matching precision. Experimental results demonstrate that the proposed algorithm outperforms the classical methods with lower computation time.