Visual Relationship Detection with Language Priors

TL;DR

This paper introduces a scalable visual relationship detection model that leverages language priors from word embeddings to predict numerous relationships in images, improving object localization and content-based image retrieval.

Contribution

The authors propose a novel approach combining object and predicate models with language priors, enabling scalable prediction of thousands of relationships with limited training data.

Findings

Outperforms previous models in relationship prediction accuracy

Can predict thousands of relationships using few examples

Enhances image retrieval through relationship understanding

Abstract

Visual relationships capture a wide variety of interactions between pairs of objects in images (e.g. "man riding bicycle" and "man pushing bicycle"). Consequently, the set of possible relationships is extremely large and it is difficult to obtain sufficient training examples for all possible relationships. Because of this limitation, previous work on visual relationship detection has concentrated on predicting only a handful of relationships. Though most relationships are infrequent, their objects (e.g. "man" and "bicycle") and predicates (e.g. "riding" and "pushing") independently occur more frequently. We propose a model that uses this insight to train visual models for objects and predicates individually and later combines them together to predict multiple relationships per image. We improve on prior work by leveraging language priors from semantic word embeddings to finetune the likelihood of a predicted relationship. Our model can scale to predict thousands of types of relationships from a few examples. Additionally, we localize the objects in the predicted relationships as bounding boxes in the image. We further demonstrate that understanding relationships can improve content based image retrieval.