Learning To Hash With Semantic Similarity Metrics And Empirical KL Divergence

Abstract

Learning to hash is an efficient paradigm for exact and approximate nearest neighbor search from massive databases. Binary hash codes are typically extracted from an image by rounding output features from a CNN, which is trained on a supervised binary similar/ dissimilar task. Drawbacks of this approach are: (i) resulting codes do not necessarily capture semantic similarity of the input data (ii) rounding results in information loss, manifesting as decreased retrieval performance and (iii) Using only class-wise similarity as a target can lead to trivial solutions, simply encoding classifier outputs rather than learning more intricate relations, which is not detected by most performance metrics. We overcome (i) via a novel loss function encouraging the relative hash code distances of learned features to match those derived from their targets. We address (ii) via a differentiable estimate of the KL divergence between network outputs and a binary target distribution, resulting in minimal

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyarponen2020learning

Related papers

• Compact Hash Code Learning With Binary Deep Neural Network (2017)9.03
• Deep Class-wise Hashing: Semantics-preserving Hashing Via Class-wise Loss (2018)11.19
• Discriminative Supervised Hashing For Cross-modal Similarity Search (2018)7.81
• A Survey On Learning To Hash (2016)21.62
• Learning To Hash With Binary Deep Neural Network (2016)14.93
• Dual Asymmetric Deep Hashing Learning (2018)9.03
• Binary Constrained Deep Hashing Network For Image Retrieval Without Manual Annotation (2018)5.84
• Deep Hashing: A Joint Approach For Image Signature Learning (2016)5.24