In this paper

we propose a new multiple arbitrary shape regions-of-interest(ROIs) coding framework.Gray mask is incorporated to manifest the priorities of the multiple ROIs.Multiple ROIs are transformed sequentially by the ISA-DWT according to their priorities with different constraints of the target bitrate and the reconstructed regional fidelity of different priorities.Then the set partitioning in hierarchical trees(SPIHT) algorithm is modified to generate the separate compression bit streams.Experimental results show that the proposed coding framework has the following novelties:(1) it ensures the quality of the reconstructed ROIs according to their priorities;(2) it is fit for arbitrary shape ROI loss and lossless coding;(3) it achieves low complexity by reaching the target ROI PSNR.