Error-resilient image and video coding over lossy networks

The rapid growth of the Internet and increasing bandwidth in the past decade has heightened the interest in robust transmission of image and video signals over error-prone links. Recent literatures highlight the multiple description (MD) coding as a promising approach to solve this problem. MD coding is a source coding technique that provides a graceful quality degradation in the presence of packet losses during transmission. Topics of interest to construct MD coders with high coding efficiency and robustness are studied in this thesis.; To minimize the overall distortion of reconstructed image, we propose an edge-adaptive estimation algorithm that is able to capture the arbitrarily-oriented edges around the corrupted coefficients. It reconstructs the lost samples more accurately compared to the previously proposed methods in the literatures. A feature-oriented MD coding scheme is proposed by adopting this estimation algorithm to smartly identify the coefficients that are sensitive to packet loss. The redundancy is then inserted to each channel by jointly optimizing the subsets of sensitive coefficients and the related quantizers. The corrupted samples are reconstructed with the proposed estimation algorithm at the receiver.; By introducing the extra prediction loops to the correlating transform based MD video coder, the mismatching error, i.e. the drift, incurred by the one-channel reconstruction is decreased. However, the problem of effective redundancy allocation is introduced by the drift coding. An efficient algorithm is developed to solve this problem. The proposed approach optimally adjusts the source coding rates for the correlating transform coefficients and drifts in a coordinated fashion.; The proposed algorithms are implemented based on the image/video MD coding systems using DCT or wavelet transforms. The advantages of the introduced approaches to achieve high coding efficiency and error-resilience for transmission are demonstrated.