Detecting Visual Relationships with Deep Relational Networks
This addresses the problem of image understanding for computer vision applications, representing an incremental advance by improving upon existing classification-based methods.
The paper tackles the problem of detecting visual relationships among objects in images, which is challenging due to high diversity and large numbers of categories, and proposes an integrated framework with a Deep Relational Network that achieves substantial improvement over state-of-the-art on two large datasets.
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.