How a General-Purpose Commonsense Ontology can Improve Performance of Learning-Based Image Retrieval
This work addresses visual reasoning tasks for AI systems by integrating rule-based knowledge sources, though it is incremental as it builds on existing deep learning methods with a filtering step.
The paper tackled the problem of sentence-based image retrieval by incorporating a general-purpose commonsense ontology (ConceptNet) into state-of-the-art vision systems, showing that it improves performance on a common benchmark dataset when filtered for visually relevant relations.
The knowledge representation community has built general-purpose ontologies which contain large amounts of commonsense knowledge over relevant aspects of the world, including useful visual information, e.g.: "a ball is used by a football player", "a tennis player is located at a tennis court". Current state-of-the-art approaches for visual recognition do not exploit these rule-based knowledge sources. Instead, they learn recognition models directly from training examples. In this paper, we study how general-purpose ontologies---specifically, MIT's ConceptNet ontology---can improve the performance of state-of-the-art vision systems. As a testbed, we tackle the problem of sentence-based image retrieval. Our retrieval approach incorporates knowledge from ConceptNet on top of a large pool of object detectors derived from a deep learning technique. In our experiments, we show that ConceptNet can improve performance on a common benchmark dataset. Key to our performance is the use of the ESPGAME dataset to select visually relevant relations from ConceptNet. Consequently, a main conclusion of this work is that general-purpose commonsense ontologies improve performance on visual reasoning tasks when properly filtered to select meaningful visual relations.