On Support Relations and Semantic Scene Graphs

TL;DR

This paper introduces a novel method for inferring support relations and constructing semantic scene graphs in scene understanding, leveraging physical stability and prior knowledge, resulting in more accurate scene graphs without pixel-wise labeling.

Contribution

The paper proposes a new approach to infer support relations using physical stability and prior knowledge, improving accuracy over existing methods and enabling scene graph construction without pixel-wise labels.

Findings

Support relations inferred are more accurate than state-of-the-art methods.

Scene graphs constructed show higher fidelity compared to ground truth.

Method evaluated successfully on NYUv2 database.

Abstract

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.