H.264-based multiple description coding using motion compensated temporal interpolation

Abstract

Abstract-Multiple description coding is a framework adapted to noisy transmission environments. In this work, we use H.264 to create two descriptions of a video sequence, each of them assuring a minimum quality level. If both of them are received, a suitable algorithm is used to produce an improved quality sequence. The key technique is a temporal image interpolation using motion compensation, inspired to the distributed video coding context. The interpolated image blocks are weighted with the received blocks obtained from the other description. The optimal weights are computed at the encoder and efficiently sent to the decoder as side information. The proposed technique shows a remarkable gain for central decoding with respect to similar methods available in the state of the art.

Low-Latency Video Streaming With Congestion Control in Mobile Ad-Hoc Networks

“…How the use of a decoding strategy reflects on the video quality depends on the codec used and the concealment strategy employed; in our experiments, we used an MDC technique that we introduced in [7], employing frame freezing as concealment. In Fig.…”

“…Several MDC techniques have been proposed, in particular for video [5]. A first class of solutions, called channel splitting techniques, are based on a suitable sampling of the original signal (e.g., temporal sampling: odd images are encoded in the first description and even images in the second one) and an associated method for recovering missing samples [6,7]; other techniques include non-conventional quantisers [8,9], progressive coding with unequal protection [10], redundant and correlating transforms [11] (which enable scalability in the case of wavelets [12]), or insertion of redundant pictures to reduce error propagation [13]. However, we refer the reader to the excellent survey papers by Goyal [4] and by Wang et al [5] for further information.…”

An MDC-based video streaming architecture for mobile networks

“…In the reference technique [8], the sequence of weights is computed at the encoder to minimise the distortion ( ) between the block in the original frame and the convex combination, and the resulting rate ( ) for the sequence depends on its conditional entropy given the quantised context. We hereby propose the following improvement: first, we estimate ( | ( )) with a preliminary test; then, the conditional entropy is used as an estimation of the coding cost˜ ( ) and, for each block, we find the optimal value * , defined as:…”

“…the one proposed in [8]) we used as legacy coder the H.264/AVC reference software JM [22], version 17.0. We selected a set of QPs (22,25,28,31,33, and 36) in order to compare the RD performance of the two methods.…”

“…Within this context, we recently proposed an MDC scheme for double description coding which is entirely based on pre-and postprocessing [8]. In this work, we propose an improved version of this scheme, and integrate it into a cross-layer framework for video transmission on wireless ad-hoc networks [9].…”