Multiple descriptions based on multirate coding for JPEG 2000 and H.264/AVC.

Area of Science:

• Multimedia data compression and transmission
• Information theory and coding
• Computer networks and distributed systems

Background:

• Multiple Description Coding (MDC) enhances data resiliency using redundant representations.
• Current MDC methods aim for quality independent of the specific subset of received descriptions.
• Efficient redundancy exploitation and network adaptability are key challenges in MDC.

Purpose of the Study:

• To propose an MDC method that ensures decoding quality depends solely on the number of received descriptions.
• To achieve efficient redundancy exploitation and adaptability to diverse network conditions.
• To apply the proposed MDC principle to JPEG 2000 images and H.264/AVC video.

Main Methods:

• Encoding source data at multiple rates.
• Blending data encoded at different rates to generate descriptions.
• Applying the method to JPEG 2000 images and H.264/AVC video within application-layer overlays.
• Fine-tuning encoder parameters for network adaptability.

Main Results:

• The proposed method demonstrates efficient redundancy exploitation.
• The technique shows easy adaptation to different network scenarios.
• Experimental results indicate favorable comparisons with state-of-the-art MDC techniques.
• The method ensures decoding quality is dependent only on the quantity of received descriptions.

Conclusions:

• The proposed MDC method offers an efficient and adaptable solution for resilient multimedia distribution.
• The technique shows significant advantages over existing MDC approaches, particularly in complex network topologies.
• This work contributes to improving the reliability of multimedia content delivery over networks.