Modeling Visual Compatibility through Hierarchical Mid-level Elements

Abstract

In this paper we present a hierarchical method to discover mid-level elements with the objective of modeling visual compatibility between objects. At the base-level, our method identifies patterns of CNN activations with the aim of modeling different variations/styles in which objects of the classes of interest may occur. At the top-level, the proposed method discovers patterns of co-occurring activations of base-level elements that define visual compatibility between pairs of object classes. Experiments on the massive Amazon dataset show the strength of our method at describing object classes and the characteristics that drive the compatibility between them.

29 pages, 19 Figures

References in corpus (4)

• Caffe: Convolutional Architecture for Fast Feature Embedding
• Inferring Networks of Substitutable and Complementary Products
• Learning Visual Clothing Style with Heterogeneous Dyadic Co-occurrences
• Mining Mid-level Visual Patterns with Deep CNN Activations

Cited by in corpus (5)

• Improving Outfit Recommendation with Co-supervision of Fashion Generation
• Visual Explanation by Interpretation: Improving Visual Feedback Capabilities of Deep Neural Networks
• Regularizing Reasons for Outfit Evaluation with Gradient Penalty
• Recommending Outfits from Personal Closet
• Mining Mid-level Visual Patterns with Deep CNN Activations