On Support Relations and Semantic Scene Graphs
This work addresses scene understanding in computer vision and photogrammetry, offering an incremental improvement by enhancing support relation accuracy without requiring pixel-wise semantic labeling.
The paper tackles the problem of inferring support relations and constructing semantic scene graphs for scene understanding by incorporating physical stability and prior support knowledge, achieving more precise support relations than state-of-the-art methods as evaluated on the NYUv2 database.
Scene understanding is a popular and challenging topic in both computer vision and photogrammetry. Scene graph provides rich information for such scene understanding. This paper presents a novel approach to infer such relations and then to construct the scene graph. Support relations are estimated by considering important, previously ignored information: the physical stability and the prior support knowledge between object classes. In contrast to previous methods for extracting support relations, the proposed approach generates more accurate results, and does not require a pixel-wise semantic labeling of the scene. The semantic scene graph which describes all the contextual relations within the scene is constructed using this information. To evaluate the accuracy of these graphs, multiple different measures are formulated. The proposed algorithms are evaluated using the NYUv2 database. The results demonstrate that the inferred support relations are more precise than state-of-the-art. The scene graphs are compared against ground truth graphs.