Least squares quantization in PCM
IEEE Transactions on Information Theory
Fully automatic clustering system
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
We are interested in the vector quantization problem. Many researches focus on finding a codebook using spatial distance. In order to evaluate the codebook, distortion is usually used. In this work, we introduce a novel concept for generating a codebook. Density estimation plays an important role to create codewords combining with a cluster distance. A vector quantization algorithm is proposed in this paper, some experimental results show that the output codebook can express the density distribution of the original vectors. Finally, three-dimensional vector problem is also considered.