Analytical model-based bit allocation for wavelet coding with applications to multiple description coding and region of interest coding

We address the problem of allocating bits to the different regions in an image coded with a progressive wavelet coder such as SPIHT (Set Partitioning in Hierarchical Trees) [1]. This type of problem appears in applications such as Region of Interest (ROI) coding or Multiple Description Coding (MDC). The wavelet coefficients in both cases are divided by different factors before coding to enable different bit allocation to different regions, because the coefficients in each region are refined at different speeds. While this is a popular approach for ROI coding [2, 3], we propose using it for MDC as well. In this work, we introduce a priority scaling factor ( ) as a dividing factor. The main contribution of this work is to provide an analytical technique to determine what the should be, given criteria such as relative importance of the regions in ROI coding or degree of redundancy in an MDC. Our approach is based on an approximation to Mallat’s model [4]. We show how our selection of is basically the same as that obtained by optimization of empirical data, with significantly less complexity.