Vector Quantization by Minimizing Kullback-Leibler Divergence
This addresses vector quantization for image classification, but appears incremental as it adapts an existing divergence measure to this task.
The paper tackles the problem of vector quantization by minimizing Kullback-Leibler Divergence to preserve class label information, resulting in a new iterative algorithm evaluated on image classification.
This paper proposes a new method for vector quantization by minimizing the Kullback-Leibler Divergence between the class label distributions over the quantization inputs, which are original vectors, and the output, which is the quantization subsets of the vector set. In this way, the vector quantization output can keep as much information of the class label as possible. An objective function is constructed and we also developed an iterative algorithm to minimize it. The new method is evaluated on bag-of-features based image classification problem.