SHREWD: Semantic Hierarchy-based Relational Embeddings For Weakly-supervised Deep Hashing

Abstract

Using class labels to represent class similarity is a typical approach to training deep hashing systems for retrieval; samples from the same or different classes take binary 1 or 0 similarity values. This similarity does not model the full rich knowledge of semantic relations that may be present between data points. In this work we build upon the idea of using semantic hierarchies to form distance metrics between all available sample labels; for example cat to dog has a smaller distance than cat to guitar. We combine this type of semantic distance into a loss function to promote similar distances between the deep neural network embeddings. We also introduce an empirical Kullback-Leibler divergence loss term to promote binarization and uniformity of the embeddings. We test the resulting SHREWD method and demonstrate improvements in hierarchical retrieval scores using compact, binary hash codes instead of real valued ones, and show that in a weakly supervised hashing setting we are able

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyarponen2019shrewd

Related papers

• Deep Class-wise Hashing: Semantics-preserving Hashing Via Class-wise Loss (2018)11.19
• Supervised Deep Hashing For Hierarchical Labeled Data (2017)8.09
• Semantic Hierarchy Preserving Deep Hashing For Large-scale Image Retrieval (2019)5.24
• SSDH: Semi-supervised Deep Hashing For Large Scale Image Retrieval (2016)15.40
• Deep Hashing With Semantic Hash Centers For Image Retrieval (2025)2.26
• Deep Discrete Hashing With Self-supervised Pairwise Labels (2017)9.49
• Hierarchy Neighborhood Discriminative Hashing For An Unified View Of Single-label And Multi-label Image Retrieval (2019)0.00
• Weakly Supervised Deep Image Hashing Through Tag Embeddings (2018)11.29