Image matching based on SIFT features and kd-tree
Minjie Li, Liqiang Wang, Ying Hao · 2010
This paper uses the specific SIFT (Scale Invariant Feature Transform) feature and a fast kd-tree matching strategy combined with invariant feature to solve single face recognition problems. It transforms face image data into scale-invariant coordinates relative to local features. The similarity was computed using Euclidean distance and kd-tree search. Experiments show that the features are invariant to image scaling, translation, rotation, and change in viewpoint.