作者: Viswanathan Swaminathan , Rashmi Mittal
DOI:
关键词:
摘要: Techniques are disclosed for the improvement of vector quantization (VQ) codebook generation. The improved codebooks may be used compression in cloud-based video applications. VQ achieves by vectorizing input streams, matching those vectors to entries, and replacing them with indexes matched along residual represent difference between stream vector. combination index is generally smaller than which they collectively encode, thus providing compression. generated from training streams grouping together similar types data (e.g., image data, motion control data) generate longer having higher dimensions greater structure. This improves ability remove redundancy increase efficiency. Storage space reduced transmission faster.