Star Student Project An image is
represented as a Bag-of-Words (BoW) histogram, where each word bin
counts Star Student Project how many local features of this image
fall in the corresponding feature space partition Star Student
Project of this word. To this end, many vector quantization
schemes are proposed to build visual vocabulary, such as K-means Star
Student Project, Hierarchical K-means (Vocabulary Tree) ,
Approximate K-means and their variances Star Student Project.
Meanwhile, hashing local features into a discrete set of bins and
indexed subsequently is an alternative choice, for which methods like
Locality Star Student Project Sensitive Hashing (LSH),
Kernalized LSH Star Student Project, and Spectral Hashing
are also exploited in the literature.
