Detecting Visual Relationships with Deep Relational Networks

TL;DR

This paper introduces Deep Relational Networks, a novel deep learning framework that effectively models object relationships in images, significantly improving visual relationship detection over previous methods.

Contribution

The paper presents Deep Relational Networks, a new approach that explicitly models statistical dependencies between objects and their relationships for better image understanding.

Findings

Achieves substantial improvement over state-of-the-art on large datasets.

Effectively models complex object relationships.

Outperforms previous classification-based methods.

Abstract

Relationships among objects play a crucial role in image understanding. Despite the great success of deep learning techniques in recognizing individual objects, reasoning about the relationships among objects remains a challenging task. Previous methods often treat this as a classification problem, considering each type of relationship (e.g. "ride") or each distinct visual phrase (e.g. "person-ride-horse") as a category. Such approaches are faced with significant difficulties caused by the high diversity of visual appearance for each kind of relationships or the large number of distinct visual phrases. We propose an integrated framework to tackle this problem. At the heart of this framework is the Deep Relational Network, a novel formulation designed specifically for exploiting the statistical dependencies between objects and their relationships. On two large datasets, the proposed method achieves substantial improvement over state-of-the-art.