Filters








755 Hits in 3.6 sec

Unsupervised Semantic Hashing with Pairwise Reconstruction [article]

Casper Hansen and Christian Hansen and Jakob Grue Simonsen and Stephen Alstrup and Christina Lioma
2020 pre-print
Inspired by this, we present Semantic Hashing with Pairwise Reconstruction (PairRec), which is a discrete variational autoencoder based hashing model.  ...  hash codes (i.e., pairwise reconstruction).  ...  PAIRWISE RECONSTRUCTION BASED HASHING Pairwise reconstruction based hashing (PairRec) is a discrete variational autoencoder with a pairwise reconstruction loss.  ... 
doi:10.1145/3397271.3401220 arXiv:2007.00380v1 fatcat:rwbqme7tdffvzhon6twmujhwmq

Deep Semantic-Preserving Reconstruction Hashing for Unsupervised Cross-Modal Retrieval

Shuli Cheng, Liejun Wang, Anyu Du
2020 Entropy  
In order to further solve the problem of unsupervised cross-modal retrieval semantic reconstruction, we propose a novel deep semantic-preserving reconstruction hashing (DSPRH).  ...  Deep hashing is the mainstream algorithm for large-scale cross-modal retrieval due to its high retrieval speed and low storage capacity, but the problem of reconstruction of modal semantic information  ...  In unsupervised cross-modal hashing, Su et al. proposed [12] a semantic reconstruction framework to mine modal semantic information.  ... 
doi:10.3390/e22111266 pmid:33287034 pmcid:PMC7712897 fatcat:akxumxcwijfxpphu523bfechee

Pairwise Supervised Hashing with Bernoulli Variational Auto-Encoder and Self-Control Gradient Estimator [article]

Siamak Zamani Dadaneh, Shahin Boluki, Mingzhang Yin, Mingyuan Zhou, Xiaoning Qian
2020 arXiv   pre-print
We propose a pairwise loss function with discrete latent VAE to reward within-class similarity and between-class dissimilarity for supervised hashing.  ...  Semantic hashing has become a crucial component of fast similarity search in many large-scale information retrieval systems, in particular, for text data.  ...  Architecture for Semantic Hashing (NASH-DN-S) Table 2 : 2 The performances of different unsupervised hash- ing models on the Reuters dataset with different lengths of hashing codes.  ... 
arXiv:2005.10477v1 fatcat:gnau24adebddbc7qltwy3jstme

Deep Unsupervised Hashing for Large-Scale Cross-Modal Retrieval Using Knowledge Distillation Model

Mingyong Li, Qiqi Li, Lirong Tang, Shuang Peng, Yan Ma, Degang Yang, Syed Hassan Ahmed
2021 Computational Intelligence and Neuroscience  
To solve this problem, inspired by knowledge distillation, we propose a novel unsupervised knowledge distillation cross-modal hashing method based on semantic alignment (SAKDH), which can reconstruct the  ...  Specifically, firstly, the teacher model adopted an unsupervised semantic alignment hashing method, which can construct a modal fusion similarity matrix.  ...  , especially when compared with supervised methods. e main reason is the lack of pairwise similarity knowledge of training data pairs. e output of unsupervised models usually contains some inaccurate semantic  ... 
doi:10.1155/2021/5107034 pmid:34326867 pmcid:PMC8310450 fatcat:i5egnylkangxdjqio3tj3exxk4

A Survey on Deep Hashing Methods

Xiao Luo, Haixin Wang, Daqing Wu, Chong Chen, Minghua Deng, Jianqiang Huang, Xian-Sheng Hua
2022 ACM Transactions on Knowledge Discovery from Data  
Moreover, deep unsupervised hashing is categorized into similarity reconstruction-based methods, pseudo-label-based methods and prediction-free self-supervised learning-based methods based on their semantic  ...  In this survey, we detailedly investigate current deep hashing algorithms including deep supervised hashing and deep unsupervised hashing.  ...  We also thank Zeyu Ma, Huasong Zhong and Xiaokang Chen who discussed with us and provided instructive suggestions.  ... 
doi:10.1145/3532624 fatcat:7lxtu2qzvvhrpnjngefli2mvca

SADIH: Semantic-Aware DIscrete Hashing

Zheng Zhang, Guo-sen Xie, Yang Li, Sheng Li, Zi Huang
2019 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Meanwhile, a semantic-aware autoencoder is developed to jointly preserve the data structures in the discriminative latent semantic space and perform data reconstruction.  ...  Specifically, a semantic-aware latent embedding is introduced to asymmetrically preserve the full pairwise similarities while skillfully handle the cumbersome n×n pairwise similarity matrix.  ...  Due to the absence of semantic label information, unsupervised hashing is usually inferior to supervised hashing, which can produce state-of-the-art retrieval results.  ... 
doi:10.1609/aaai.v33i01.33015853 fatcat:6ypjbpi2nbcqbpf3anfpfmhts4

Adaptive Structural Similarity Preserving for Unsupervised Cross Modal Hashing [article]

Liang Li, Baihua Zheng, Weiwei Sun
2022 arXiv   pre-print
We introduce structural semantic metrics based on graph adjacency relations during the semantic reconstruction and correlation mining stage and meanwhile align the structure semantics in the hash space  ...  In addition, most of them mainly focus on association mining and alignment among pairwise instances in continuous space but ignore the latent structural correlations contained in the semantic hashing space  ...  Pairwise Instance Structural Similarity Reconstruction Similarity Alignment Correlation Preservation The meadow is peopled with wild flowers.  ... 
arXiv:2207.04214v1 fatcat:47735my6qfe3pagk3hv65w64ue

SADIH: Semantic-Aware DIscrete Hashing [article]

Zheng Zhang, Guo-sen Xie, Yang Li, Sheng Li, Zi Huang
2019 arXiv   pre-print
Meanwhile, a semantic-aware autoencoder is developed to jointly preserve the data structures in the discriminative latent semantic space and perform data reconstruction.  ...  Specifically, a semantic-aware latent embedding is introduced to asymmetrically preserve the full pairwise similarities while skillfully handle the cumbersome n times n pairwise similarity matrix.  ...  Due to the absence of semantic label information, unsupervised hashing is usually inferior to supervised hashing, which can produce state-of-the-art retrieval results.  ... 
arXiv:1904.01739v2 fatcat:lh2la3lngzf3dlki7jfwdd2hr4

A Survey on Deep Hashing Methods [article]

Xiao Luo, Haixin Wang, Daqing Wu, Chong Chen, Minghua Deng, Jianqiang Huang, Xian-Sheng Hua
2022 arXiv   pre-print
Moreover, deep unsupervised hashing is categorized into similarity reconstruction-based methods, pseudo-label-based methods and prediction-free self-supervised learning-based methods based on their semantic  ...  In this survey, we detailedly investigate current deep hashing algorithms including deep supervised hashing and deep unsupervised hashing.  ...  We also thank Zeyu Ma, Huasong Zhong and Xiaokang Chen who discussed with us and provided instructive suggestions.  ... 
arXiv:2003.03369v5 fatcat:m2iu3htilvgztkcazw3cyk6iqe

Locality preserving hashing

Yi-Hsuan Tsai, Ming-Hsuan Yang
2014 2014 IEEE International Conference on Image Processing (ICIP)  
Furthermore, pairwise label similarity can be further incorporated in the weight matrix to bridge the semantic gap between data and hash codes.  ...  The spectral hashing algorithm relaxes and solves an objective function for generating hash codes such that data similarity is preserved in the Hamming space.  ...  Other methods such as binary reconstructive embeddings (BRE) [15] and minimal loss hashing [16] aim to learn hash codes by minimizing the reconstruction errors between the semantic space and Hamming  ... 
doi:10.1109/icip.2014.7025604 dblp:conf/icip/TsaiY14 fatcat:gm4aolnjzjfcpotvm3n5v7fisi

Binary Constrained Deep Hashing Network for Image Retrieval without Manual Annotation [article]

Thanh-Toan Do, Tuan Hoang, Dang-Khoa Le Tan, Trung Pham, Huu Le, Ngai-Man Cheung, Ian Reid
2018 arXiv   pre-print
In particular, to deal with the non-smoothness of binary constraints, we propose a novel pairwise constrained loss function, which simultaneously encodes the distances between pairs of hash codes, and  ...  However, training deep hashing networks for the task is challenging due to the binary constraints on the hash codes, the similarity preserving property, and the requirement for a vast amount of labelled  ...  Comparison with unsupervised hashing methods 4.2.1 Comparison with traditional unsupervised hashing methods All compared traditional unsupervised methods require image features as inputs, instead of raw  ... 
arXiv:1802.07437v7 fatcat:aoyjsntydrbofaye4ucwh56nv4

Deep Hashing Based on VAE-GAN for Efficient Similarity Retrieval

Guoqing Jin, Yongdong Zhang, Ke Lu
2019 Chinese journal of electronics  
The method combines a Variational autoencoder (VAE) with a Generative adversarial network (GAN) to generate content preserving images for pairwise hashing learning.  ...  By accepting real image and systhesized image in a pairwise form, a semantic perserving feature mapping model is learned under a adversarial generative process.  ...  The semantic perserving hashing network simultaneously optimizes the pairwise cross-entropy loss on semantic similarity pairs and the pairwise quantization loss on compact hash codes.  ... 
doi:10.1049/cje.2019.08.001 fatcat:4hll6wzxcfcqhpzfbmqmusrbue

Deep Self-Adaptive Hashing for Image Retrieval [article]

Qinghong Lin, Xiaojun Chen, Qin Zhang, Shangxuan Tian, Yudong Chen
2021 arXiv   pre-print
However, most deep unsupervised hashing methods usually pre-compute a similarity matrix to model the pairwise relationship in the pre-trained feature space.  ...  To solve the aforementioned problems, we propose a Deep Self-Adaptive Hashing (DSAH) model to adaptively capture the semantic information with two special designs: Adaptive Neighbor Discovery (AND) and  ...  With the lack of labels, most of the unsupervised deep hashing methods construct a pairwise similarity matrix with pre-trained deep features. For example, Yang et al.  ... 
arXiv:2108.07094v2 fatcat:nzwrowa6dbbgnmbta6v537inly

Unsupervised Deep Pairwise Hashing

Ye Ma, Qin Li, Xiaoshuang Shi, Zhenhua Guo
2022 Electronics  
Extensive experiments on large-scale benchmark databases demonstrate the effectiveness of the proposed method, which outperforms recent state-of-the-art unsupervised hashing methods with significantly  ...  To alleviate this issue, in this paper, we propose a novel unsupervised deep pairwise hashing method to effectively and robustly exploit the similarity information between training samples and multiple  ...  Unsupervised hashing with a binary deep neural network (UH-BDNN) [12] utilizes the reconstruction loss to encourage the similarity among samples.  ... 
doi:10.3390/electronics11050744 fatcat:mr2s6g3lhrevpitvzjhqt6egxq

Deep Semantic Hashing Using Pairwise Labels

Richeng Xuan, Junho Shim, Sang-goo Lee
2021 IEEE Access  
Pairwise label data have not been employed in previous VAE-based hashing methods. In a recent study using pairwise information [11] , pairwise generation was employed for unsupervised learning.  ...  We selected the following unsupervised baselines for comparison: locality sensitive hashing (LSH) 3 , spectral hashing (SpH) 4 [17] , Self-Taught Hashing (STH) [12] , variational deep semantic hashing  ... 
doi:10.1109/access.2021.3092150 fatcat:5at4wt7cobfhbi4ngwlv3s6ynm
« Previous Showing results 1 — 15 out of 755 results