Techniques for improving the performance of distributed video coding

Distributed Video Coding (DVC) is a recently proposed paradigm in video communication, which fits well emerging applications such as wireless video surveillance, multimedia sensor networks, wireless PC cameras, and mobile cameras phones. These applications require a low complexity encoding, while possibly affording a high complexity decoding. DVC presents several advantages: First, the complexity can be distributed between the encoder and the decoder. Second, the DVC is robust to errors, since it uses a channel code. In DVC, a Side Information (SI) is estimated at the decoder, using the available decoded frames, and used for the decoding and reconstruction of other frames. In this Ph.D thesis, we propose new techniques in order to improve the quality of the SI. First, successive refinement of the SI is performed after each decoded DCT band, using a Partially Decoded WZF (PDWZF along with the reference frames. Moreover, in this refinement approach an adaptive search area algorithm is also proposed, that allows adapting the search area to the current motion between the WZF and the reference frames, using the PDWZF obtained after decoding the first DCT band. Then, a new scheme for SI generation based on backward, forward motion estimations, and Quad-tree refinement is proposed. Furthermore, in the aim of enhancing the quality of the decoded WZFs for larger GOP sizes, an algorithm based on adjacent decoded frames is investigated, using an adaptive search area and a variable block size. Another contribution of this thesis concerns a fusion of global and local SI. Global parameters are estimated at the encoder using the Scale-Invariant Feature Transform (SIFT) algorithm. These global parameters are sent to the decoder to estimate the global SI. Then, new methods for combining global and local motion estimations are proposed, to further improve the SI. In the first approach, the differences between the corresponding blocks are used to combine the global and local SI frames. In the second approach, Support Vector Machine (SVM) is used to combine the two SI frames. In addition, algorithms are proposed to refine the fusion during the decoding process by exploiting the PDWZF and the decoded DC coefficients. Furthermore, the foreground objects are used in the combination of the global and local motion estimations, using elastic curves and foreground objects motion compensation. Extensive experiments have been conducted showing that important gains are obtained by the proposed techniques compared to the classical DISCOVER codec. In addition, the performance of DVC applying the proposed algorithms outperforms now the performance of H.264/AVC Intra and H.264/AVC No motion for tested sequences. Besides that, the gap with H.264/AVC in an Inter IB…IB configuration is significantly reduced.