检索结果-内蒙古大学图书馆

FaiResGAN: Fair and robust blind face restoration with biometrics preservation

Image and Vision Computing 2025年 160卷

作者： George Azzopardi Antonio Greco Mario Vento Bernoulli Institute for Mathematics Computer Science and Artificial Intelligence University of Groningen The Netherlands Department of Information and Electrical Engineering and Applied Mathematics (DIEM) University of Salerno Italy National Research Council of Italy Institute for High Performance Computing and Networking Italy

Modern computer vision technologies enable systems to detect, recognize, and analyze facial features, but challenges arise when images are noisy, blurred, or low quality. Blind face restoration, which aims to recover high-quality facial images without prior knowledge of degradation, addresses this issue. In this paper, we introduce Fair Restoration GAN (FaiResGAN), a novel Generative Adversarial Network (GAN) designed to balance face restoration with the preservation of soft biometrics (identity, ethnicity, age, and gender). Our model incorporates a pseudo-random batch composition algorithm to promote fairness and mitigate bias, alongside a realistic degradation model simulating corruptions typical in surveillance images. Experimental results show that FaiResGAN outperforms state-of-the-art blind face restoration methods, both quantitatively and qualitatively. A user study involving 40 participants showed that FaiResGAN-restored images were preferred by 70% of users. Additionally, tests on VGGFace2, UTKFace, and FairFace datasets demonstrate FaiResGAN’s superior performance in preserving soft biometric attributes and ensuring fair restoration across different genders and ethnicities.

关键词： Blind face restoration Facial soft biometrics Face analysis Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

A method of generating random bits by using electronic bipolar memristor

引用

Chinese Physics B 2020年第4期29卷 535-540页

作者： Bin-Bin Yang Nuo Xu Er-Rui Zhou Zhi-Wei Li Cheng Li Pin-Yun Yi Liang Fang Institute for Quantum Information&State Key Laboratory of High Performance Computing College of ComputerNational University of Defense TechnologyChangsha 410073China College of Computer National University of Defense TechnologyChangsha 410073China College of Electronic Science and Engineering National University of Defense TechnologyChangsha 410073China

The intrinsic stochasticity of resistance switching process is one of the holdblocks for using memristor as a fundamental element in the next-generation nonvolatile ***,such a weakness can be used as an asset for generating the random bits,which is valuable in a hardware security *** this work,a forming-free electronic bipolar Pt/Ti/Ta2O5/Pt memristor is successfully fabricated to investigate the merits of generating random bits in such a *** resistance switching mechanism of the fabricated device is ascribed to the electric field conducted electrons trapping/de-trapping in the deep-energy-level traps produced by the"oxygen grabbing"*** stochasticity of the electrons trapping/detrapping governs the random distribution of the set/reset switching voltages of the device,making a single memristor act as a random bit in which the resistance of the device represents information and the applied voltage pulse serves as the triggering *** physical implementation of such a random process provides a method of generating the random bits based on memristors in hardware security applications.

关键词： memristor resistance switching electrons trapping/de-trapping random bits

来源：评论

学校读者我要写书评

暂无评论

Research on Effectiveness of Contrastive Learning with Adversarial Examples for Different Downstream Tasks 3

Research on Effectiveness of Contrastive Learning with Adver...

引用

3rd IEEE International Conference on Frontiers Technology of Information and computer, ICFTIC 2021

作者： Zhang, Bin Sun, Yuxia Chen, Zhao Luo, Liang Xu, Yingmei Jinan University Guangdong Provincial Key Laboratory of High-Performance Computing Department of Computer Science Guangzhou China Jinan University Department of Computer Science Guangzhou China Guangdong Science Technology Infrastructure Center Guangdong Provincial Key Laboratory of High-Performance Computing Guangzhou China

ISBN: (纸本)9781665406055

Contrastive learning is a new self-supervised representation learning technique, which is considered to have great potential to improve the performance of downstream learning tasks. Recently, some researchers have conducted experiments on some datasets and found that contrastive learning based on adversarial examples can improve the performance of some image classification tasks. However, does contrastive learning with adversarial examples really benefit the performance of various downstream tasks? At present, there is still a lack of in-depth research on this issue. In this paper, we introduce a contrastive learning method with adversarial examples based on different adversarial perturbations intensities, and establish a variety of downstream image classification tasks. Experiments on different types of downstream tasks show that contrastive learning based on adversarial examples may not improve the performance of downstream tasks. We also find that the adversarial perturbation intensity and the proportion of the adversarial perturbation loss in contrastive loss are important factors affecting the above performance. © 2021 IEEE.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

How can network coding help P2P content distribution?

How can network coding help P2P content distribution?

引用

2009 IEEE International Conference on Communications, ICC 2009

作者： Ma, Guanjun Xu, Yinlong Kaiqian, Ou Wen, Luo Department of Computer Science and Technology University of Science and Technology of China Key Laboratory on High Performance Computing Anhui Province China

ISBN: (纸本)9781424434350

It is well known that network coding can enhance the performance of Peer-to-Peer (P2P) content distribution systems since it benefits block scheduling. In this paper, we introduce our P2P content distribution system "SmartCode" with sparse network coding on PlanetLab. SmartCode uses pre-checking to avoid linearly dependent blocks being transmitted. Under the same system architecture, we also implement two systems "Ecode" and "LRF". Ecode is also based on sparse network coding, but without pre-checking. LRF is BitTorrent-like, using local-rarest-first for block scheduling. We conduct extensive experiments to compare the performances among the three systems. Experimental results show that the distributing time of SmartCode is reduced by 11% on average and up to 19% compared with LRF, and by 7% on average and up to 16% compared with Ecode. We also conduct experiments to analyze robustness among the three systems under the assumption that peers join in and leave a content distribution session dynamically and the seed leave the session after it has transmitted a fixed percentage of blocks over the total blocks of the content. In dynamical cases, SmartCode outperforms evidently than Ecode and LRF in downloading time and the times of peers completing the downloading. ©2009 IEEE.

关键词： Network coding

来源：评论

学校读者我要写书评

暂无评论

A study of code reuse and sharing characteristics of Java applications 1

A study of code reuse and sharing characteristics of Java ap...

引用

1st Workshop on Workload Characterization, WWC 1998

作者： Conte, M.T. Trick, A.R. Gyllenhaal, J.C. Hwu, W.W. Department of Electrical and Computer Engineering Coordinated Science Laboratory Center for Reliable and High-performance Computing University of Illinois at Urbana-Champaign United States

ISBN: (纸本)0769504507

Presents a detailed characterization of Java application and applet workloads in terms of reuse and sharing of Java code at the program, class and method level. In order to expose more sharing opportunities, techniques for detecting code equivalence even in the presence of minor code changes or constant pool index differences) are also proposed and examined. The analyzed application workload consists of the recently released SPECjvm98 benchmarks, and the applet workload is derived from three extensive searches of the Internet between May 1997 and May 1998 using an enhanced Web crawler. Analysis of these workloads reveals several new code sharing and optimization opportunities. © 1998 IEEE.

关键词： Java programming language

来源：评论

学校读者我要写书评

暂无评论

Multicalorics II

引用

Journal of Applied Physics 2025年第19期137卷

作者： Wosnitza, J. Slaughter, J. Zhang, Q.M. Hou, H. Dresden High Magnetic Field Laboratory (HLD-EMFL) Helmholtz-Zentrum Dresden-Rossendorf and Würzburg-Dresden Cluster of Excellence ct.qmat Dresden 01328 Germany Institut für Festkörper- und Materialphysik Technische Universität Dresden Dresden 01062 Germany Ames National Laboratory U.S. Department of Energy Iowa State University Ames 50011 IA United States School of Electrical Engineering and Computer Science Materials Research Institute The Pennsylvania State University University Park 16802 PA United States School of Materials Science and Engineering Beihang University Beijing 100191 China Tianmushan Laboratory Zhejiang Provincial Laboratory for Aviation Hangzhou 311312 China

来源：评论

学校读者我要写书评

暂无评论

Dasher: A peer-to-peer content distribution system based on combined network coding

Dasher: A peer-to-peer content distribution system based on ...

引用

2009 2nd IEEE International Conference on Broadband Network and Multimedia Technology, IEEE IC-BNMT2009

作者： Kaiqian, Ou Yinlong, Xu Guanjun, Ma Yulin, Zhu Department of Computer Science and Technology University of Science and Technology of China Key Laboratory on High Performance Computing Anhui Province China

ISBN: (纸本)9781424450053

Network coding can ease the block scheduling and thus makes the distribution more efficient. However, the complexities of encoding and decoding increase sharply as the content size scales up. In this paper, we propose a coding scheme which combines chunked coding and sparse linear network coding to reduce both encoding and decoding costs of content distribution. Based on the combined scheme, we implement a P2P content distribution system, named Dasher, where Local-Rarest-First scheme is adopted for chunk scheduling. Under the same system architecture, we implement three comparative systems, a BitTorrent-like system, named Mybt, a system only with sparse coding, named Sparser and a system using chunked coding, named Chunker. We conduct extensive experiments to compare the performance among these four systems. The experimental results show that Dasher with certain chunk sizes can reduce the average downloading time up to 15% compared with Mybt, and up to 43% with Chunker. With proper chunk sizes, the downloading time of Dasher is almost the same with Sparser. The average decoding rate of Dasher is the same with Chunker, and is nearly m times as fast as Sparser, where m is the number of chunks. Moreover, with respect to robustness, Dasher performs almost as well as Chunker, better than Mybt, but worse than Sparser. ©2009 IEEE.

关键词： Network coding

来源：评论

学校读者我要写书评

暂无评论

Lexicographically fair bandwidth allocation in multi-session layered media multicast using network coding

Lexicographically fair bandwidth allocation in multi-session...

引用

2009 2nd IEEE International Conference on Broadband Network and Multimedia Technology, IEEE IC-BNMT2009

作者： Xiaobin, Lin Yinlong, Xu Yulin, Zhu Kaiqian, Ou Department of Computer Science and Technology University of Science and Technology of China Key Laboratory on High Performance Computing Anhui Province China

ISBN: (纸本)9781424450053

Layering techniques effectively cope with network heterogeneity encountered in media distribution. Recent work on network coding shows that the multicast throughput can be greatly improved by allowing coding operation at the intermediate nodes in the network. In this paper, we study the problem of lexicographically fair bandwidth allocation to make the receivers of different sessions share network bandwidth fairly in multi-session layered media multicast using network coding and formulate it into an integer linear programming (ILP). Because the ILP is not time efficient when the network is large, a LP-based polynomial time algorithm ApprOpt is proposed to approximate the optimal solution of the ILP. Numerical results show that the throughput of all media multicast sessions is significantly improved by network coding, and that algorithm ApprOpt allocates network bandwidth very close to the optimal solution of the ILP and utilizes network bandwidth efficiently. ©2009 IEEE.

关键词： Network coding

来源：评论

学校读者我要写书评

暂无评论

Validity checking on grid service composition

Validity checking on grid service composition

引用

31st Annual International computer Software and Applications Conference, COMPSAC 2007

作者： Zhou, Jing Zeng, Guosun Department of Computer Science and Technology Tongji University Shanghai 201804 China Tongji Branch National Engineering and Technology Center of High Performance Computer

ISBN: (纸本)9780769528700

Grid service is composed of elementary services and the execution procedure is characterized by dynamical evolvement and exceptional no-determinism. The existing service description language, BPEL4WS (abbreviated to BPEL), only provides static service description, which cannot guarantee the correct composition of elementary services and dynamically monitor the occurrence of the exceptional events. According to the problems mentioned above, a verification approach to check the validity of the behavior partial order sequence of a grid service is proposed in this paper, which converts the grid service script based on BPEL into the behavior sequence described in Cπ-calculus automatically, and extracts the partial order constraint rules on the behavior of elementary services in a grid service, based on which to check the reachability of elementary services and validity of the behavior sequence during the execution of a grid service. A validity check algorithm is proposed finally. © 2007 IEEE.

关键词： Grid computing

来源：评论

学校读者我要写书评

暂无评论

Using unified parallel C to parallelize the FDTD method of computational electrodynamics

Using unified parallel C to parallelize the FDTD method of c...

引用

2009 IEEE 13th International Multitopic Conference, INMIC 2009

作者： Hussain, A. Shafi, A. Raza, J. Center for High Performance Scientific Computing Department of Computing NUST School of Electrical Engineering and Computer Science Pakistan

ISBN: (纸本)9781424448722

This paper discusses a parallel implementation of the Finite-Difference Time-Domain (FDTD) method using Unified Parallel C (UPC). UPC is an extension of ANSI C - primarily designed for programming distributed memory parallel hardware. The FDTD method is a powerful and widely used numerical technique in computational electromagnetics community. It is a resource - CPU cycles and memory - hungry technique. Consequently, a parallel implementation is highly desirable to decrease time-to-solution and be able to study larger problems with limited resources. To the best of our knowledge, this is the first attempt to exploit UPC to parallelize the FDTD method. We evaluate the performance of our implementation on a cluster of four SunFire v890 compute nodes-each equipped with 16 processing cores - and report impressive speedups. ©2009 IEEE.

关键词： Finite difference time domain method

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：