Now showing 1 - 6 of 6
Thumbnail Image
Publication

Skip Prediction and Early Termination for Fast Mode Decision in H.264/AVC

2006-08-31, Ivanov, Yuri, Bleakley, Chris J.

This paper proposes a fast mode decision algorithm for H.264/AVC based on skip prediction and early termination techniques. The skip decision is based on a partially computed Sum of Absolute Differences metric combined with utilization of the Lagrangian Rate- Distortion cost function from the previous frame. A statistical analysis of the spatio-temporal characteristics of the Rate Distortion cost function and SAD metric for skip decisions is provided. Experimental results show that the new algorithm outperforms existing ones in performance providing a 55% reduction in total computational complexity, a slight reduction in bit rate and negligible impact on visual quality.

Thumbnail Image
Publication

Fast Mode Decision With Early Termination For H.264/AVC Video Coding

2006-10-28, Ivanov, Yuri, Bleakley, Chris J.

In this paper a fast mode decision algorithm based on an early termination procedure is proposed for H.264/AVC video encoding. Unlike previous methods, the termination decision is based on the rate distortion cost function. A statistical analysis of the spatio-temporal characteristics of the rate distortion cost function and of the probability of mode transition is given for test sequences. Experimental results show that the new algorithm provides a 38% reduction in total computational complexity with a negligible increase in the bit rate and negligible reduction in visual quality when compared to conventional encoding.

Thumbnail Image
Publication

Pareto-Optimal Macroblock Classification for Fast Mode Decision in H.264

2007-11-27, Ivanov, Yuri, Bleakley, Chris J.

This paper presents a novel fast mode decision algorithm for H.264/AVC based on a Pareto-optimal macroblock classification strategy. Previously published H.264 low complexity schemes mostly concentrated on improving class decision metrics, but did not justify the choice of MD classes. Herein, we use Pareto analysis to derive the optimal set of MD classes and to define efficient class decision metrics. For each MD class only rate-distortion optimal complexity settings are used. Experimental results show that the proposed algorithm outperforms previously published algorithms, providing a 57-73% reduction in total computational complexity with some reduction in bit rate and acceptable visual quality.

Thumbnail Image
Publication

Survey and Pareto Analysis Method for Coding Efficiency Assessment of Low Complexity H.264 Algorithms

2006-09-01, Ivanov, Yuri, Bleakley, Chris J.

A large number of algorithms have been proposed by researchers to reduce H.264computational complexity. Currently there is no method for reliably comparing theeffectiveness of these algorithms. This paper proposes a method that allows direct comparison of the results obtained for various previously published low complexity H.264 encoding algorithms. The method is based on a new coding efficiency metric for unified bit rate and quality assessment. Pareto analysis is used to derive an optimal reference efficiency complexity curve using standard H.264 encoding tools and parameters. The paper demonstrates application of the method to the assessment of recently published low-complexity algorithms. The method shows that some published low complexity algorithms can be outperformed by simply adjusting the standard video encoder parameters.

Thumbnail Image
Publication

Adaptive Lagrange Multiplier for Low Bit Rates in H.264

2005-08-31, Ivanov, Yuri, Bleakley, Chris J.

This paper describes a novel adaptive Lagrange multiplier algorithm based on the general Rate-Distortion model with dynamic lambda calculation for H.264 video coding. When experimentally tested with the JM 8.1a encoder on QCIF video sequences the algorithm provides 2.8% average bit rate saving with insignificant PSNR loss (around 0.13%) without computational complexity increase. Experiments indicate that the algorithm may also be adopted for higher bit rates.

Thumbnail Image
Publication

Dynamic Complexity Scaling for Real-Time H.264/AVC Video Encoding

2007-09-28, Ivanov, Yuri, Bleakley, Chris J.

The H.264 video encoding standard can achieve high coding efficiency at the expense of high computational complexity. Typically, real-time software implementation requires omission of most optional encoding tools leading to significantly reduced coding efficiency. This paper proposes a novel method for real-time H.264 encoding based on dynamic control of the encoding parameters to meet real-time constraints while minimizing coding efficiency loss. Experimental results show that the method provides up to 19% lower bit rate than conventional real-time encoding using fixed parameters with the same visual quality. The method allows real-time 30fps QCIF encoding on a Pentium IV with similar coding efficiency to full search baseline profile encoding.