Rate distortion optimized quantization for H.264/AVC based on dynamic programming

Wensheng Wang; Huijuan Cui; Kun Tang

doi:10.1117/12.633421

24 June 2005 Rate distortion optimized quantization for H.264/AVC based on dynamic programming

Wensheng Wang, Huijuan Cui, Kun Tang

Proceedings Volume 5960, Visual Communications and Image Processing 2005; 596065 (2005) https://doi.org/10.1117/12.633421
Event: Visual Communications and Image Processing 2005, 2005, Beijing, China

Abstract

In this paper, an optimal quantization decision algorithm in the rate distortion sense for H.264/AVC1 with dynamic programming is proposed, in which a trellis is constructed for the 4x4 block quantization and adaptive entropy encoding process. Dynamic programming is employed to go through the trellis and search for the best path with minimum rate distortion cost in linear time. The quantization decisions for all coefficients in one block are made simultaneously in our approach. Thorough experiments are conducted on various sequences with different coding complexity for a wide range of bit-rates. The results show that the performance gain increases as the reproduction PSNR gets higher. For the sequences with high coding complexity, up to approximate 1 dB gain over independent quantization is observed for high PSNR encoding, e.g. higher than 45dB. In other cases, obvious gain can also be achieved. At the same time, since the state space is compact, the complexity increase is acceptable for software implementation.

Citation Download Citation

Wensheng Wang, Huijuan Cui, and Kun Tang "Rate distortion optimized quantization for H.264/AVC based on dynamic programming", Proc. SPIE 5960, Visual Communications and Image Processing 2005, 596065 (24 June 2005); https://doi.org/10.1117/12.633421

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available