Deep Hashing Network for Efficient Similarity Retrieval

By Han Zhu, Mingsheng Long, Jianmin Wang. AAAI Conference on Artificial Intelligence, 2016

682 citations recorded by the source corpus.

Due to the storage and retrieval efficiency, hashing has been widely deployed to approximate nearest neighbor search for large-scale multimedia retrieval. Supervised hashing, which improves the quality of hash coding by exploiting the semantic similarity on data pairs, has received increasing attention recently. For most existing supervised hashing methods for image retrieval, an image is first represented as a vector of hand-crafted or machine-learned features, followed by another separate quantization step that generates binary codes. However, suboptimal hash coding may be produced, because the quantization error is not statistically minimized and the feature representation is not optimally compatible with the binary coding. In this paper, we propose a novel Deep Hashing Network (DHN) architecture for supervised hashing, in which we jointly learn good image representation tailored to hash coding and formally control the quantization error. The DHN model constitutes four key components: (1) a sub-network with multiple convolution-pooling layers to capture image representations; (2) a fully-connected hashing layer to generate compact binary hash codes; (3) a pairwise cross-entropy loss layer for similarity-preserving learning; and (4) a pairwise quantization loss for controlling hashing quality. Extensive experiments on standard image retrieval datasets show the proposed DHN model yields substantial boosts over latest state-of-the-art hashing methods.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/eb0c64244dcf238a2cbf479ab2fdc9047fc80bc5

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

HashNet: Deep Learning to Hash by Continuation (2017, methodology): “Recently, deep learning to hash methods [37, 20, 31, 8, 40] have shown that end-to-end learning of feature representation and hash coding can be more effective using deep neural networks [18, 2], which can naturally encode any nonlinear hash functions.”
Binary Generative Adversarial Networks for Image Retrieval (2017, methodology): “We follow the settings in [49] and use the subset of 195,834 images from the 21 most frequent concepts, where each concept consists of at least 5,000 images.”
Deep Multi-View Enhancement Hashing for Image Retrieval (2020, methodology): “[24] proposed the DHN which uses a pairwise cross-entropy loss L for similarity-preserving learning and a pairwise quantization loss Q for controlling hashing quality.”
Deep Visual-Semantic Hashing for Cross-Modal Retrieval (2016, background): “Meanwhile, latest deep hashing methods [40, 23, 46, 5] yielded state of art results on many datasets, but these methods are limited to single-modal retrieval.”
Deep Cauchy Hashing for Hamming Space Retrieval (2018, methodology): “We follow protocol in [36, 3] to randomly select 100 images per class as query set, 500 images per class as training set, and the rest images are used as the database.”
Deep Supervised Discrete Hashing (2017, methodology): “Besides the triplet ranking based methods, some pairwise label based deep hashing methods are also exploited [9, 27].”
A Decade Survey of Content Based Image Retrieval Using Deep Learning (2020, background): “representation for hash codes and controlling the quantization error [73].”
Deep Sketch Hashing: Fast Free-Hand Sketch-Based Image Retrieval (2017, methodology): “Related Work Hashing techniques [42, 16, 34, 62, 39, 30, 55, 35, 17, 14, 74, 36, 14, 70, 37, 47, 38, 54, 25, 33, 41] have recently been successfully applied to encode highdimensional features into compact similarity-preserving binary codes, which enables extremely fast similarity search by the use o”
A Survey on Deep Transfer Learning (2018, background): “[30] learning domain adaptation and deep hash features simultaneously in a DNN.”
Efficient Passage Retrieval with Hashing for Open-domain Question Answering (2021, methodology): “Recently, many deep-learning-to-hash methods have been proposed (Lai et al., 2015; Zhu et al., 2016; Li et al., 2016; Cao et al., 2017, 2018) to jointly learn feature representations and hash functions in an end-to-end manner.”
Efficient Interaction-based Neural Ranking with Locality Sensitive Hashing (2019, methodology): “Our work is motivated by the rich history of research on approximation with hashing such as LSH [1–4, 7, 8, 13, 15, 37] and in various data-intensive applications [6, 10, 17, 18, 30].”
Central Similarity Quantization for Efficient Image and Video Retrieval (2019, methodology): “The “deep learning to hash” methods such as CNNH [44], DNNH [14], DHN [48], DCH [1] and HashNet [2] have been successfully applied to image hashing.”
Deep Hashing Network for Unsupervised Domain Adaptation (2017, methodology): “The hash-fc8 provides greater stability when fine-tuning the learning rates than the deep hashing networks [30, 50].”