Journal article
Deep reinforcement hashing with redundancy elimination for effective image retrieval
Pattern recognition, Vol.100, p.107116
04/2020
DOI: 10.1016/j.patcog.2019.107116
Abstract
•Block-wise Hash Code Inference is utilized to preserve arbitrarily large global similarity relationship.•Hash Code Mapping based on Multi-binary Classification is established to be trained in a point-wise style.•Hash Bits De-redundancy based on Deep Reinforcement Learning is created to eliminate redundant or even harmful bits from hash codes while preserving the retrieval accuracy.
Hashing is one of the most promising techniques in approximate nearest neighbor search due to its time efficiency and low cost in memory. Recently, with the help of deep learning, deep supervised hashing can perform representation learning and compact hash code learning jointly in an end-to-end style, and obtains better retrieval accuracy compared to non-deep methods. However, most deep hashing methods are trained with a pair-wise loss or triplet loss in a mini-batch style, which makes them inefficient at data sampling and cannot preserve the global similarity information. Besides that, many existing methods generate hash codes with redundant or even harmful bits, which is a waste of space and may lower the retrieval accuracy. In this paper, we propose a novel deep reinforcement hashing model with redundancy elimination called Deep Reinforcement De-Redundancy Hashing (DRDH), which can fully exploit large-scale similarity information and eliminate redundant hash bits with deep reinforcement learning. DRDH conducts hash code inference in a block-wise style, and uses Deep Q Network (DQN) to eliminate redundant bits. Very promising results have been achieved on four public datasets, i.e., CIFAR-10, NUS-WIDE, MS-COCO, and Open-Images-V4, which demonstrate that our method can generate highly compact hash codes and yield better retrieval performance than those of state-of-the-art methods.
Details
- Title: Subtitle
- Deep reinforcement hashing with redundancy elimination for effective image retrieval
- Creators
- Juexu Yang - Fudan UniversityYuejie Zhang - Fudan UniversityRui Feng - Fudan UniversityTao Zhang - Shanghai University of Finance and EconomicsWeiguo Fan - University of Iowa
- Resource Type
- Journal article
- Publication Details
- Pattern recognition, Vol.100, p.107116
- Publisher
- Elsevier Ltd
- DOI
- 10.1016/j.patcog.2019.107116
- ISSN
- 0031-3203
- eISSN
- 1873-5142
- Grant note
- DOI: 10.13039/501100003399, name: Science and Technology Commission of Shanghai Municipality; DOI: 10.13039/100007219, name: Natural Science Foundation of Shanghai; DOI: 10.13039/501100001809, name: National Natural Science Foundation of China; DOI: 10.13039/501100013139, name: Humanities and Social Science Fund of Ministry of Education of China
- Language
- English
- Date published
- 04/2020
- Academic Unit
- Business Analytics
- Record Identifier
- 9984380546502771
Metrics
21 Record Views