Vector quantization and signal compression
Vector quantization and signal compression
Tree Structured Vector Quantization with Dynamic Path Search
ICPP '99 Proceedings of the 1999 International Workshops on Parallel Processing
Pattern Recognition Letters
Fast tree-structured nearest neighbor encoding for vector quantization
IEEE Transactions on Image Processing
A fast search algorithm for vector quantization using L2-norm pyramid of codewords
IEEE Transactions on Image Processing
Successive refinement lattice vector quantization
IEEE Transactions on Image Processing
An efficient encoding algorithm for vector quantization based on subvector technique
IEEE Transactions on Image Processing
Fast full search equivalent encoding algorithms for image compression using vector quantization
IEEE Transactions on Image Processing
Diagonal axes method (DAM): a fast search algorithm for vector quantization
IEEE Transactions on Circuits and Systems for Video Technology
An efficient computation of Euclidean distances using approximated look-up table
IEEE Transactions on Circuits and Systems for Video Technology
An efficient Euclidean distance computation for vector quantization using a truncated look-up table
IEEE Transactions on Circuits and Systems for Video Technology
Efficient Fingercode Classification
IEICE - Transactions on Information and Systems
A fast VQ codebook generation algorithm via pattern reduction
Pattern Recognition Letters
Vector quantization using the firefly algorithm for image compression
Expert Systems with Applications: An International Journal
Accelerating visual categorization with the GPU
ECCV'10 Proceedings of the 11th European conference on Trends and Topics in Computer Vision - Volume Part II
Hi-index | 0.10 |
Tree-structured vector quantization (TSVQ) is a highly efficient technique for locating an appropriate codeword for each input vector. The algorithm does not guarantee that the selected codeword is the closest one to the input vector. Consequently, the image quality of TSVQ is worse than that of full-search VQ (FSVQ). Although researchers have proposed multipath TSVQ and DP-TSVQ to enhance the image quality, these methods are still too slow for achieving high image quality. Therefore, this study presents a novel full search equivalent TSVQ (FSE-TSVQ) to obtain efficiently the closest codeword for each input vector. FSE-TSVQ employs the triangle inequality to achieve efficient pruning of impossible codewords. Moreover, this study also develops the enhanced DP-TSVQ (EDP-TSVQ) algorithm, which achieves a better trade-off than DP-TSVQ between encoding time and image quality. EDP-TSVQ is a hybrid technique which adds DP-TSVQ's critical function to FSE-TSVQ. EDP-TSVQ always provides an image quality identical to that of DP-TSVQ, but by searching fewer codebook tree nodes. Simulation results reveal that FSE-TSVQ requires only 21-38% of the running time of FSVQ. For a high image quality application, the performance of EDP-TSVQ is always better than that of DP-TSVQ. Using the example of a codebook tree with 512 codewords, with the threshold of the critical function set to 0.6, simulation results indicate that EDP-TSVQ requires only 37% of the execution time of DP-TSVQ.