Deep Priority Hashing
2018 · Zhangjie Cao, Ziping Sun, Mingsheng Long, et al.
Abstract
Deep hashing enables image retrieval by end-to-end learning of deep representations and hash codes from training data with pairwise similarity information. Subject to the distribution skewness underlying the similarity information, most existing deep hashing methods may underperform for imbalanced data due to misspecified loss functions. This paper presents Deep Priority Hashing (DPH), an end-to-end architecture that generates compact and balanced hash codes in a Bayesian learning framework. The main idea is to reshape the standard cross-entropy loss for similarity-preserving learning such that it down-weighs the loss associated to highly-confident pairs. This idea leads to a novel priority cross-entropy loss, which prioritizes the training on uncertain pairs over confident pairs. Also, we propose another priority quantization loss, which prioritizes hard-to-quantize examples for generation of nearly lossless hash codes. Extensive experiments demonstrate that DPH can generate high-qual
Authors
(none)
Tags
Stats
Related papers
- Deep Residual Hashing (2016)0.00
- Deep Hashing: A Joint Approach For Image Signature Learning (2016)5.24
- Improved Deep Hashing With Soft Pairwise Similarity For Multi-label Image Retrieval (2018)16.82
- Deep Class-wise Hashing: Semantics-preserving Hashing Via Class-wise Loss (2018)11.19
- Deep Discrete Hashing With Self-supervised Pairwise Labels (2017)9.49
- Query-adaptive Image Retrieval By Deep Weighted Hashing (2016)12.68
- Deep Supervised Discrete Hashing (2017)0.00
- HHF: Hashing-guided Hinge Function For Deep Hashing Retrieval (2021)14.24