Progressive edge-preserving depth maps coding based on sparse representation

Multiview Video plus Depth coding has received much attention in recent years because of its importance in many different fields, ranging from video games to medical imaging. An efficient coding, that causes the least possible distortion without excessive rate and complexity increase, is crucial particularly for depth maps. In this paper, we propose a coding method that deals with progressive depth maps rendering. Depth images are approximated by a compact representation of sparse coefficients and discrete cosine/B-spline dictionary atoms. To enhance tradeoff between sparsity gain and sharp edges preservation, proposed coding is parameterized using high and low sparsity criteria. Then, selected atoms are structured in layers such that they can be progressively communicated to the decoder side according to consumers capabilities, target applications requirements and transmission channel capacities. Experimental results show the perfomance of the proposed method as compared to some state-of-the-art algorithms.