Automatic Multi-image Photo-texturing of Complex 3d Scenes

The paper presents an approach for projective texture mapping from photographs onto triangulated surfaces from 3D laser scanning. By these means, the effort to generate photo-realistic models of complex shaped objects can be reduced considerably. The images are collected from multiple viewpoints, which do not necessarily correspond to the viewpoints of LIDAR data collection. In order to handle the resulting problem of occlusions, the visibility of the model areas in the respective images has to be established. For this purpose, the algorithm works in both image and object space and efficiently detects ambient, back-face and view frustum occlusions. Occluding polygons are labelled and separated with their connectivity to texture them recursively using the optimal of the available images until the final textured model is produced. After this visibility processing, colour values will be correctly assigned from the photograph to the visible polygons. In order to gain a high quality texture, lens distortion and colour corrections are applied during processing. The approach is demonstrated by generating high realistic 3D textured models for the Alkasneh monument in Petra city and a Romanian Theatre in the ancient city of Jerash.