Weakly-supervised learning of visual relations

Abstract

This paper introduces a novel approach for modeling visual relations between pairs of objects. We call relation a triplet of the form (subject, predicate, object) where the predicate is typically a preposition (eg. 'under', 'in front of') or a verb ('hold', 'ride') that links a pair of objects (subject, object). Learning such relations is challenging as the objects have different spatial configurations and appearances depending on the relation in which they occur. Another major challenge comes from the difficulty to get annotations, especially at box-level, for all possible triplets, which makes both learning and evaluation difficult. The contributions of this paper are threefold. First, we design strong yet flexible visual features that encode the appearance and spatial configuration for pairs of objects. Second, we propose a weakly-supervised discriminative clustering model to learn relations from image-level labels only. Third we introduce a new challenging dataset of unusual relations (UnRel) together with an exhaustive annotation, that enables accurate evaluation of visual relation retrieval. We show experimentally that our model results in state-of-the-art results on the visual relationship dataset significantly improving performance on previously unseen relations (zero-shot learning), and confirm this observation on our newly introduced UnRel dataset.

References in corpus (8)

• Rich feature hierarchies for accurate object detection and semantic segmentation
• Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations
• Generation and Comprehension of Unambiguous Object Descriptions
• Visual Relationship Detection with Language Priors
• From Captions to Visual Concepts and Back
• Weakly Supervised Action Labeling in Videos Under Ordering Constraints
• Minding the Gaps for Block Frank-Wolfe Optimization of Structured SVMs
• Deep Compositional Captioning: Describing Novel Object Categories without Paired Training Data

Cited by in corpus (4)

• Zoom-Net: Mining Deep Feature Interactions for Visual Relationship Recognition
• Graph R-CNN for Scene Graph Generation
• An Interpretable Model for Scene Graph Generation
• Shuffle-Then-Assemble: Learning Object-Agnostic Visual Relationship Features