Learning Decorrelated Hashing Codes With Label Relaxation for Multimodal Retrieval

Tian, Dayong; Wei, Yiwen; Zhou, Deyun

doi:10.1109/access.2020.2988923

Cited by 5 publications

(2 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Methods based on minimizing quantization errors apply orthogonality and balanced regularizations on the intermediate real matrices of hashing code matrices. However, it has been proven that quantization will break the orthogonality and balance except for some extremely ideal cases [14].…”

Section: A Hashing Methodsmentioning

confidence: 99%

Optimizing Non-Differentiable Metrics for Hashing

et al. 2021

Self Cite

View full text Add to dashboard Cite

Image hashing embeds the image to binary codes which can boost the efficiency of approximately nearest neighbors search. F-measure is a widely-used metric for evaluating the performance of hashing methods. However, it is non-differentiable and hence it has not been used as an object function for hashing. Heuristic algorithms, e.g. evolutionary computation and particle swarm optimization (PSO), are good at optimizing non-differentiable objectives, while they are inefficient in very high-dimensional variables which are commonly used in hashing models. To address this contradict, we propose a scheme to bridge hashing methods and F-measure objective using PSO. The hashing methods are used to generate real-valued codes for images and then the parameters of quantization procedure are optimized by PSO. Our scheme can incorporate a wide range of hashing methods, heuristic optimization algorithms and nondifferentiable metrics. Experimental results demonstrate that our scheme can be used to further improve the performance of existing hashing methods.

show abstract

Section: A Hashing Methodsmentioning

confidence: 99%

Optimizing Non-Differentiable Metrics for Hashing

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…Preventive measures rely on solid-phase methods that can go as far as allowing customers to upload identity copies until a profile is developed [30,31]. This form of methodology is referred to as a Sybil prevention strategy.…”

Section: Related Workmentioning

confidence: 99%

Trust management and data protection for online social networks

et al. 2022

View full text Add to dashboard Cite

Online social networks (OSNs) have been an integral aspect of daily life. Most people use OSNs such as Facebook and Twitter to communicate and exchange information. Numerous analytical methods have been applied to OSNs by active exploration of their graph structure, like clustering analyses for automatic online group detection and node impact analyses of awareness of influential nodes within social networks. On OSN, a large‐scale attack was launched by creating a bogus profile and using it to propagate spam, malware, and phishing attacks. To overcome the above issues, we designed a trust management and data protection model (TM‐A‐DPM), in which the trust management identifies the trust factor accuracy to achieve confidentiality, integrity, and privacy. Experimental results determine the performance and computational efficiency as well as resolve the privacy threats such as data leakage and data forgery.

show abstract