Optimised feature map finite-state vector quantisation for image coding

An optimised feature may finite-state vector quantisation (referred to as o ptimised FMFSVQ) is presented for image coding. Based on the block-based gr adient descent search algorithm used for motion estimation in video coding, the optimised FMFSVQ system finds a neighbourhood-based optimal codevector for each input vector by extending the associated state codebook stage by stage, thus rendering each slate quantiser a variable rate vector quantisat ion. The optimised FMFSVQ system can be interpreted as a cascade of a finit e-state vector quantiser and classified vector quantisers. Furthermore, an adaptive optimised FMFSVQ is obtained. Experiments demonstrate the superior rate-distortion performance of the adaptive optimised FMFSVQ compared with the original adaptive FMFSVQ and the memoryless vector quantisation.