Improving Information Extraction from Images with Learned Semantic Models
This work addresses the need for more detailed image understanding in applications, but it appears incremental as it builds on prior models without reporting concrete performance gains.
The paper tackles the problem of extracting semantic information from images by focusing on relationships between objects, comparing a previous visual-semantic model with a new conditional multi-way model for visual relationship detection.
Many applications require an understanding of an image that goes beyond the simple detection and classification of its objects. In particular, a great deal of semantic information is carried in the relationships between objects. We have previously shown that the combination of a visual model and a statistical semantic prior model can improve on the task of mapping images to their associated scene description. In this paper, we review the model and compare it to a novel conditional multi-way model for visual relationship detection, which does not include an explicitly trained visual prior model. We also discuss potential relationships between the proposed methods and memory models of the human brain.