TIME COST COMPARISON AT DIFFERENT
PHASES Star Student Project OF THE emphasizing the uniform
distribution of their visual words. Star Student Project
RELATED WORK Visual Vocabulary Construction: Building visual
vocabulary usually resorts to unsupervised vector quantization, which
Star Student Project subdivides the local feature space into
discrete regions each corresponds to a visual word. AnStar Student
Project image is represented as a Bag-of-Words BoW histogram,
where each word bin counts how many local features of this image fall
in the corresponding feature space partition of Star Student
Project 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
Star Student Project , Approximate K-means , and
their variances
