Principle of Kd-Tree Algorithm

Principle of Kd-Tree algorithm:

http://baike.baidu.com/link?url=Fk2aYUNNDNCrL2FqwSkgZA9YJN9QBpyhP9YteLDEzcq5wL0ztbGYyWmqyQ6fSi-vCyYb9L7hoAQ17lwUM0jK-_


http://blog.sina.com.cn/s/blog_6f611c300.html

Summary
in dimension compared to Kdf611c300101bys Small (for example: K≤30), the search efficiency of the algorithm is very high, but when Kd-tree is used for indexing and searching for high-dimensional data (for example: K≥100), it faces the curse of dimension ) problem, the search efficiency will drop rapidly with the increase of dimension. Usually, in practical applications, the data we often deal with have high-dimensional characteristics. For example, in image retrieval and recognition, each image is usually represented by a vector of several hundreds of dimensions, and the local features of each feature point are represented by a high-dimensional dimensional vector to represent (eg: 128-dimensional SIFT feature).