Sparsity-based optimization of two lifting-based wavelet transforms for semi-regular mesh compression

This paper describes how to optimize two popular wavelet transforms for semi-regular meshes, using lifting scheme. The objective is to adapt multiresolution analysis to the input mesh to improve its subsequent coding. Considering either the Butterflyor the Loop-based lifting schemes, our algorithm finds at each resolution level an optimal prediction operator P such that it minimizes the L1-norm of the wavelet coefficients. The update operator U is then recomputed in order to take into account the modifications to P. Experimental results show that our algorithm improves on state-of-the-art wavelet coders.