FEATURE-BASED TEXTURE SYNTHESIS AND HIERARCHICAL TENSOR APPROXIMATION BY

Texture synthesis has always been an interesting research topic in Graphics. Neighborhood-based algorithms have two common stages: search for most similar neighborhoods in the sample texture; merge a local neighborhood into the (partially) synthesized output texture. When the first stage cannot find good neighbors, the second stage may generate seams. We propose to extract a feature map from the input texture and synthesize a feature map for output. We de-velop novel algorithms to perform feature matching and alignment. This approach significantly reduces feature discontinuities and related artifacts. For 3D surfaces, things get more complicated as continuity constraints restrict the variability of synthesized textures. We propose to relax such restrictions and decompose synthesis into two stages: feature map synthesis and Laplacian texture reconstruction. Experiments indicate that this relaxation can produce desirable results for regular texture synthesis as well as texture mixing from multiple sources. Visual data approximation is another important issue in Graphics. We propose hierarchical tensor approximation to expose multi-scale and inhomogeneous structures in visual datasets