Multiresolution Block Sampling-Based Method for Texture Synthesis

We present an eficient multiresolution block samplingbased texture synthesis algorithm that works well for a wide vatiety of challenging textures. We consider the texture synthesis as a procedure that first estimates and then resumples a probability function. To achieve the eflect of good and fast estimation of the probubility function for a given texture us well us to retain the structural information duting synthesis, we propose a concept of block sampling and a corresponding novel scheme of texture synthesis bused on multiresolution block sampling (MBS). The computational complexity of the proposed ulgotithm is much lower than that of existing algorithms. Expetimentul results have shown that this novel texture synthesis ulgotithm produces supetior results for many challenging textures.