Graphical Contrastive Losses for Scene Graph Parsing

• Published in: Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online) pp. 11527 - 11535
• Main Authors: Zhang, Ji, Shih, Kevin J., Elgammal, Ahmed, Tao, Andrew, Catanzaro, Bryan
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2019
• Subjects: Categorization; Computer vision; Detectors; Entropy; Force; Genomics; Instruments; Pattern recognition; Pipelines; Recognition: Detection; Retrieval; Scene Analysis and Understanding; Semantics; Visualization
• ISSN: 1063-6919
• DOI: 10.1109/CVPR.2019.01180

Most scene graph parsers use a two-stage pipeline to detect visual relationships: the first stage detects entities, and the second predicts the predicate for each entity pair using a softmax distribution. We find that such pipelines, trained with only a cross entropy loss over predicate classes, suf...