无序多视图分类及其有序化技术

Grouping and organizing multiple unordered images

摘要:

无序图像分类问题是多视图匹配、三维重构、图像检索等应用的基础研究。提出一种新的相似度度量准则，准确划分场景相关图像和非相关图像；接着在通用匹配算法的基础上，提出试探匹配算法和种子扩散算法2种加速策略，显著提升了无序多图匹配及分类的效率，为三维重构等应用提供了更好的基础数据。

Abstract:

Grouping unordered images is considered a fundamental task in multi-view matching，3D scene modeling，image retrieval and other visual computing applications. First a robust view-similarity measure is proposed and the images can be categorized accurately；then two speedup strategies，seed growing based grouping and tentative feature matching, are presented respectively to speed up calculating effectively，which can further provide nice data for 3D modeling.