A neural-net based rate control algorithm for MPEG encoders is introduced. It is based on the global rate-distortion model constructed by a self-organising feature map. The performances of the proposed algorithm are better than those for MPEG-4 VM5+ frame-based rate control, based on the criteria of the average number of bits per frame and PSNR
Published in:
Electronics Letters
(Volume:36
,
Issue:
12
)
Date of Publication: 8 Jun 2000