Indoor scene modeling from a single image using normal inference and edge features

We present in this paper an interactive approach for semantically modeling the indoor environment given only a single indoor image as input, without requiring access to the scene or using any additional measurements like RGBD cameras. Our key insight is that, although depth estimation from a single image is notoriously difficult, we can conveniently obtain a relatively accurate normal map, which essentially conveys a great deal of scene geometry. This enables us to model each object in a data-driven manner by representing the object as a normal-based graph and retrieving a similar model from the database by graph matching. Moreover, edge information is integrated to further improve the searching result. We hypothesize a set of sparse surface orientations for the image and further refine them in an intuitive and straightforward manner. With a small amount of simple user interaction, our approach is able to generate a plausible model of the scene. To verify the effectiveness of our proposed method, we show the modeling results on a variety of indoor images.