Abstract: Local invariant features have been widely applied in many computer vision applications and high-dimensional image feature matching is a core part of solving these problems. In this paper, it proposes a new indexing structure for the high-dimensional feature matching, which is based on the distance of the sub-vectors. The algorithm converts the feature vectors into one dimensional indexing value and only searches the features indexed by the same value in nearest neighbor searching process so that the searching speed can be greatly improved and high searching accuracy can be reached at the same time. Experimental results demonstrate the efficiency and effectiveness of the proposed methods in extensive image matching and image retrieval applications.

Key words: high dimensional feature matching, nearest neighbor searching, image retrieve, indexing sub-vector distance

摘要： 图像局部不变特征已经成功地应用在计算机视觉当中的许多领域，而如何快速有效地匹配高维图像局部特征向量是解决这类问题的关键步骤。提出了一种新的基于子向量距离索引的高维特征向量匹配算法，将高维空间中最近邻搜索问题转化为一维索引值的查找和局部搜索问题，在保证较高的搜索精度的同时大大提高了搜索速度。大量的图像匹配和图像检索实验验证了该算法的有效性。

关键词: 高维特征匹配, 最近邻搜索, 图像检索, 子向量距离索引