A Novel Multiple Description Approach to Predictive Video Coding
Multiple description coding (MDC) is a source coding technique that exploits path diversity to combat packet losses over error-prone channels. In this paper, we proposed a novel drift-free multi-state MDC method. At the encoder side, the original video is
- PDF / 859,519 Bytes
- 10 Pages / 430 x 660 pts Page_size
- 56 Downloads / 202 Views
Abstract. Multiple description coding (MDC) is a source coding technique that exploits path diversity to combat packet losses over errorprone channels. In this paper, we proposed a novel drift-free multi-state MDC method. At the encoder side, the original video is compressed into multiple independently decodable H.263 streams, each with its own coding structure and prediction process, such that if one stream is lost, the other stream can still be used to produce video with acceptable quality. At the decoder side, each description is considered as a noisy observation of the original video. A Least square-error (LSE) based merge algorithm is proposed to combine the descriptions. The experimental results show that the proposed algorithm has similar coding efficiency to [1], yet with improved error resilience. Keywords: Multiple description coding (MDC), least square-error estimator (LSE), H.263, PB Frame.
1
Introduction
Transmission of video over non guaranteed Qos network is a challenging task, especially for real-time interactive multimedia applications where re-transmission is not feasible. Multiple description coding (MDC) has emerged as a promising approach to overcome such difficult situation. The basic idea is to encode the original signal into multiple coded streams, called descriptions, and transmit them over different channels. At the destination, each description alone provides low but acceptable quality and all descriptions together lead to better quality. This structure takes advantage of the fact that error/loss occurred in different channel is independent and it provides adequate quality without requiring re-transmission of any lost packets. During the past decade, various MD compression algorithm have been proposed. Comprehensive reviews of the MD technique can be found in [2]. However, incorporating MDC for video compression is still a challenge. The reason is that all conventional video compression standards adopt a predictive coding
This work has been supported by the Innovation and Technology Commission of the Hong Kong Special Administrative Region, China (project no. GHP/033/05).
H.H.S. Ip et al. (Eds.): PCM 2007, LNCS 4810, pp. 286–295, 2007. c Springer-Verlag Berlin Heidelberg 2007
Encoder
MD Encoder
Joint Error Concealment
Decoder
287
Reconstructed Video
Decoder
Signal Composition
Encoder
Network
Signal Decomposition
Original Video
A Novel Multiple Description Approach to Predictive Video Coding
MD Decoder
Fig. 1. System Architecture of Multi-state MDC
technique and encoding the prediction error using MDC is not straightforward. Moreover, a frame that uses two descriptions, for example, have three possible reconstructions at the decoder. The encoder may use a signal for prediction that is unavailable to the decoder (due to loss of one or more descriptions) and error will occur and accumulate in the following motion-compensated frames. Such phenomenon is called drifting and it is a fundamental design concern in MD Video coders. One class of MDC methods, known as multi-state
Data Loading...
