Image retrieval using scene graphs

TLDR

The paper proposes a novel framework for semantic image retrieval using scene graphs. The framework represents objects, attributes, and relationships in scene graphs, uses them as queries, and applies a conditional random field model to score possible groundings for ranking, evaluated on a new dataset of 5,000 human‑generated scene graphs. Experiments show that using full scene graphs outperforms object‑only or low‑level feature methods, and the model also improves object localization over baselines.

Abstract

This paper develops a novel framework for semantic image retrieval based on the notion of a scene graph. Our scene graphs represent objects ("man", "boat"), attributes of objects ("boat is white") and relationships between objects ("man standing on boat"). We use these scene graphs as queries to retrieve semantically related images. To this end, we design a conditional random field model that reasons about possible groundings of scene graphs to test images. The likelihoods of these groundings are used as ranking scores for retrieval. We introduce a novel dataset of 5,000 human-generated scene graphs grounded to images and use this dataset to evaluate our method for image retrieval. In particular, we evaluate retrieval using full scene graphs and small scene subgraphs, and show that our method outperforms retrieval methods that use only objects or low-level image features. In addition, we show that our full model can be used to improve object localization compared to baseline methods.