Knowledge-Embedded Routing Network for Scene Graph Generation

TL;DR

This paper introduces a Knowledge-Embedded Routing Network that leverages statistical correlations in scene graphs to improve relationship prediction, especially for less frequent interactions, demonstrating superior performance on the Visual Genome dataset.

Contribution

The paper proposes a novel network that embeds statistical relationship knowledge via a structured graph and routing mechanism, enhancing scene graph generation accuracy.

Findings

Outperforms state-of-the-art methods on Visual Genome dataset

Effectively addresses unbalanced relationship distribution

Utilizes a structured knowledge graph for better relationship inference

Abstract

To understand a scene in depth not only involves locating/recognizing individual objects, but also requires to infer the relationships and interactions among them. However, since the distribution of real-world relationships is seriously unbalanced, existing methods perform quite poorly for the less frequent relationships. In this work, we find that the statistical correlations between object pairs and their relationships can effectively regularize semantic space and make prediction less ambiguous, and thus well address the unbalanced distribution issue. To achieve this, we incorporate these statistical correlations into deep neural networks to facilitate scene graph generation by developing a Knowledge-Embedded Routing Network. More specifically, we show that the statistical correlations between objects appearing in images and their relationships, can be explicitly represented by a structured knowledge graph, and a routing mechanism is learned to propagate messages through the graph to explore their interactions. Extensive experiments on the large-scale Visual Genome dataset demonstrate the superiority of the proposed method over current state-of-the-art competitors.