Context-aware Patch-based Method for Façade Inpainting

Abstract

Realistic representations of 3D urban scenes is an important aspect of scene understanding and has many applications. Given untextured polyhedral Level-of-Detail 2 (LoD2) models of building and imaging containing fac ¸ade textures, occlusions caused by foreground objects are an essential disturbing factor of fac ¸ade textures. We developed a modification of a well-known patch-based inpainting method and used the knowledge about fac ¸ade details in order to improve the fac ¸ade inpainting of occlusions. Our modification focuses on suppression of undesired, superfluous repetitions of textures. To achieve this, a coarse inpainting result by a structural-based method is used to influence the choice of the best patch so that homogeneous regions are preferred. The coarse inpainting is calculated using the context knowledge and average color instead of traditionally applied arbitrary structural inpainting. Our modification furthermore introduces a parameter that allows to weight the influence of the coarse inpainting. A parameter study shows that this parameter can be chosen intuitively and does not require any parameter choice method. The cleaned fac ¸ade textures could be successfully integrated into the accordingly adjusted building models thus upgrading them to LoD3.