检索结果-内蒙古大学图书馆

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Chuangchuang Tan Huan Liu Yao Zhao Shikui Wei Guanghua Gu Ping Liu Yunchao Wei Institute of Information Science Beijing Jiaotong University Beijing Key Laboratory of Advanced Information Science and Network Technology School of Information Science and Engineering Yanshan University Hebei Key Laboratory of Information Transmission and Signal Processing CSE department University of Nevada Reno USA

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

Recently, the proliferation of highly realistic synthetic images, facilitated through a variety of GANs and Diffusions, has significantly heightened the susceptibility to misuse. While the primary focus of deepfake detection has traditionally centered on the design of detection algorithms, an investigative inquiry into the generator architectures has remained conspicuously absent in recent years. This paper contributes to this lacuna by rethinking the architectures of CNN-based generator, thereby establishing a generalized representation of synthetic artifacts. Our findings illuminate that the up-sampling operator can, beyond frequency-based artifacts, produce generalized forgery artifacts. In particular, the local interdependence among image pixels caused by upsampling operators is significantly demon-strated in synthetic images generated by GAN or diffusion. Building upon this observation, we introduce the concept of Neighboring Pixel Relationships(NPR) as a means to capture and characterize the generalized structural artifacts stemming from up-sampling operations. A comprehensive analysis is conducted on an open-world dataset, comprising samples generated by 28 distinct generative models. This analysis culminates in the establishment of a novel state-of-the-art performance, showcasing a remarkable 12.8% im-provement over existing methods. The code is available at https://***/chuangchuangtan/NPR-DeepfakeDetection.

关键词： Deepfakes Computer vision Frequency-domain analysis Computer architecture Detectors Generative adversarial networks Generators

来源：评论

学校读者我要写书评

暂无评论

Learning on Gradients: Generalized Artifacts Representation for GAN-Generated Images Detection

Learning on Gradients: Generalized Artifacts Representation ...

引用

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Chuangchuang Tan Yao Zhao Shikui Wei Guanghua Gu Yunchao Wei Institute of Information Science Beijing Jiaotong University Beijing Key Laboratory of Advanced Information Science and Network Technology School of Information Science and Engineering Yanshan University Hebei Key Laboratory of Information Transmission and Signal Processing

Recently, there has been a significant advancement in image generation technology, known as GAN. It can easily generate realistic fake images, leading to an increased risk of abuse. However, most image detectors suffer from sharp performance drops in unseen domains. The key of fake image detection is to develop a generalized representation to describe the artifacts produced by generation models. In this work, we introduce a novel detection framework, named Learning on Gradients (LGrad), designed for identifying GAN-generated images, with the aim of constructing a generalized detector with cross-model and cross-data. Specifically, a pretrained CNN model is employed as a transformation model to convert images into gradients. Subsequently, we leverage these gradients to present the generalized artifacts, which are fed into the classifier to ascertain the authenticity of the images. In our framework, we turn the data-dependent problem into a transformation-model-dependent problem. To the best of our knowledge, this is the first study to utilize gradients as the representation of artifacts in GAN-generated images. Extensive experiments demonstrate the effectiveness and robustness of gradients as generalized artifact representations. Our detector achieves a new state-of-the-art performance with a remarkable gain of 11.4%. The code is released at https://***/chuangchuangtan/LGrad.

关键词：

来源：评论

学校读者我要写书评

暂无评论

First-Shot Unsupervised Anomalous Sound Detection with Unknown Anomalies Estimated by Metadata-Assisted Audio Generation

First-Shot Unsupervised Anomalous Sound Detection with Unkno...

引用

International Conference on Acoustics, Speech, and signal processing (ICASSP)

作者： Hejing Zhang Qiaoxi Zhu Jian Guan Haohe Liu Feiyang Xiao Jiantong Tian Xinhao Mei Xubo Liu Wenwu Wang Group of Intelligent Signal Processing College of Computer Science and Technology Harbin Engineering University Harbin China National Engineering Laboratory for Modeling and Emulation in E-Government Harbin Engineering University Harbin China Centre for Audio Acoustics and Vibration University of Technology Sydney Ultimo Australia Centre for Vision Speech and Signal Processing University of Surrey Guildford UK

First-shot (FS) unsupervised anomalous sound detection (ASD) is a brand-new task introduced in DCASE 2023 Challenge Task 2, where the anomalous sounds for the target machine types are unseen in training. Existing methods often rely on the availability of normal and abnormal sound data from the target machines. However, due to the lack of anomalous sound data for the target machine types, it becomes challenging when adapting the existing ASD methods to the first-shot task. In this paper, we propose a new framework for the first-shot unsupervised ASD, where metadata-assisted audio generation is used to estimate unknown anomalies, by utilising the available machine information (i.e., metadata and sound data) to fine-tune a text-to-audio generation model for generating the anomalous sounds that contain unique acoustic characteristics accounting for each different machine type. We then use the method of Time-Weighted Frequency domain audio Representation with Gaussian Mixture Model (TWFRGMM) as the backbone to achieve the first-shot unsupervised ASD. Our proposed FS-TWFR-GMM method achieves competitive performance amongst top systems in DCASE 2023 Challenge Task 2, while requiring only 1% model parameters for detection, as validated in our experiments.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-objective Deployment of WSNs in Underground Sheltered Space 19th

Multi-objective Deployment of WSNs in Underground Sheltere...

引用

19th EAI International Conference on Heterogeneous Networking for Quality, Reliability, Security and Robustness, QShine 2023

作者： Wan, Liangtian Wang, Caiyun Sun, Lu Chen, Boyu Zheng, Jibin Wang, Xianpeng Key Laboratory for Ubiquitous Network and Service Software of Liaoning Province School of Software Dalian University of Technology Dalian116620 China Department of Communication Engineering Institute of Information Science Technology Dalian Maritime University Dalian116026 China School of Communications and Information Engineering Chongqing University of Posts and Telecommunications Chongqing China National Laboratory of Radar Signal Processing Xidian University Xi’an710071 China State Key Laboratory of Marine Resource Utilization in South China Sea and School of Information and Communication Engineering Hainan University Haikou570228 China

ISBN: (纸本)9783031651250

Given the increasing population, the underground spaces are widely used. The existing research on wireless sensor networks (WSNs) development has mainly focused on overground space;also, the existing work was based on an ideal deployment environment and simple sensor behaviour. As a consequence, they do not satisfy requirements for practical 3D environment. The 3D wireless sensor networks (WSNs) deployment has new difficulties in the underground space. The signal transmission is impacted by obstacles in the underground space. Thus, we propose a geometry-based 3D signal propagation model to calculate the signal path loss caused by obstacles. Then, taking both coverage and connectivity into account, we transform the WSNs deployment problem into a multi-objective optimization problem (MOP). Finally, we compare some state-of-the-art multi-objective evolutionary algorithms (MOEAs) for the WSNs deployment problem in underground sheltered space. By the experimental results, we optimize the multiobjective deployment problem of the WSNs in the underground sheltered space. © ICST Institute for Computer sciences, Social Informatics and Telecommunications Engineering 2024.

关键词： Wireless sensor networks

来源：评论

学校读者我要写书评

暂无评论

A Speaker Recognition Method Based on Stable Learning

A Speaker Recognition Method Based on Stable Learning

引用

International Conference on Acoustics, Speech, and signal processing (ICASSP)

作者： Jian Zhang Jing Ma Xiaochen Guo Lin Li Liang He School of Computer Science and Technology Xinjiang University China Xinjiang Key Laboratory of Signal Detection and Processing China School of Electronic Science and Engineering Xiamen University China Department of Electronic Engineering Tsinghua University China

With the development of deep learning, speaker recognition systems have shown increasingly better performance. The generalization ability of the models is also an important aspect of performance evaluation. Typically, a baseline system is used to compare against the improved models to demonstrate performance enhancements. However, we cannot determine the differences in learned voiceprint features between the improved models and the baseline system. This paper introduces an improved speaker recognition system based on the ECAPA-TDNN model. It utilizes stable learning to eliminate sample correlation and employs attribution analysis to compare the differences in voiceprint feature learning between the improved and baseline systems. Experimental results demonstrate that stable learning improves the model’s generalization performance and helps it learn better voiceprint features. The effectiveness and generalization capability of the proposed method are verified through experiments on the VoxCeleb, CNCeleb, and LibriSpeech datasets. This work is important for enhancing speaker recognition performance, analyzing differences in voiceprint feature learning, and promoting advancements in the field.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-Layer Visual Perception for No-Reference Image Quality Assessment 6

Multi-Layer Visual Perception for No-Reference Image Quality...

引用

6th IEEE International Conference on Electronic Information and Communication technology, ICEICT 2023

作者： Qi, Junwei Wang, Yingzhen Wang, Qingchun Li, Yingsong College of Computer Science and Technology Harbin Engineering University Harbin150001 China College of Information and Communication Engineering Harbin Engineering University Harbin150001 China Shanghai Yanding Information Technology Co. Ltd. Shanghai201210 China Anhui University Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui Hefei230601 China

ISBN: (纸本)9798350399059

In this work, we propose a no-reference image quality evaluation approach, aiming to solve the problem that the traditional convolutional neural network is insufficient to express the global information of the image. Therefore, in order to make the captured image features have the coherence of global context information, we combine a Transformer structure widely used in NLP (natural language processing) with the traditional CNN to realize the attention mechanism in the human visual perception characteristics. Firstly, the self-attention mechanism of Transformer is used to learn the global representation of the image from the multi-layer features extracted from different levels of CNN. Additionally, to forecast the ultimate image quality score, the global and local attributes are combined. Finally, the proposed algorithm is evaluated on three public datasets commonly used in CSIQ [1], TID2013 [2] and LIVE [3]. Compared with eight traditional no-reference image quality evaluation algorithms and five methods based on deep learning, the proposed algorithm achieves competitive results. © 2023 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Optical and Electrical Analysis of an Advanced 4T All-perovskite Tandem Solar Cell

Optical and Electrical Analysis of an Advanced 4T All-perovs...

引用

Progress in Electromagnetic Research Symposium (PIERS)

作者： Zhaosheng Xia Xingang Ren Shengyang Wei Zhixiang Huang Information Materials and Intelligent Sensing Laboratory of Anhui Province Institute of Physical Science and Information Technology Anhui University Hefei China Key Laboratory of Intelligent Computing & Signal Processing Ministry of Education Anhui University Hefei China

ISBN: (数字)9798350375909

ISBN: (纸本)9798350375916

The investigation of the optoelectronic characteristics of all-perovskite tandem solar cells holds pivotal significance in surpassing the Shockley-Queisser limit of single-junction perovskite solar cells. Initially, we established two distinct four-terminal tandem solar cells (4T TSCs) composed of different perovskite materials: FA 0.8 Cs 0.2 Pb(I 0.7 Br 0.3 ) 3 (E g = 1.75 eV) and MASnI 3 (E g = 1.3 eV). Through numerical simulations of the optoelectronic characteristics of 4T TSCs, we evaluated the performance of the TSCs at varying perovskite layer thicknesses. Following optimization, the power conversion efficiency (PCE) of the 4T TSCs increased from the initial 30.34% to 33.54%. Through an in-depth analysis of the cell and energy distribution in TSCs, it was observed that the light-coupling layer significantly influenced the PCE of the 4T TSCs. Upon optimizing the light-coupling layer, the PCE of the TSC increased from the initial 30.34% to 34.7%. This study provides crucial theoretical and experimental foundations for optimizing the performance of all-perovskite TSCs.

关键词： Photovoltaic cells Numerical simulation Perovskites Optical coupling Optical materials Power conversion Electromagnetics

来源：评论

学校读者我要写书评

暂无评论

Knowledge-Aided Covariance Matrix Estimation Algorithm Based on Geometric Mean in STAP for Airborne Radar

Knowledge-Aided Covariance Matrix Estimation Algorithm Based...

引用

IET International Radar Conference 2023, IRC 2023

作者： Jing, Yang Du, Xiaolin Cui, Guolong Zheng, Jibin Li, Jianbo School of Computer and Control Engineering Yantai University Yantai China School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu China National Key Laboratory of Radar Signal Processing Xidian University Xi'an China School of Communication and Information Engineering Chongqing University of Posts and Telecommunications Chongqing China

ISBN: (纸本)9781839539954

The problem of knowledge-aided (KA) covariance matrix estimation in airborne radar space-time adaptive processing (STAP) is studied in this paper. In the traditional KA covariance matrix estimation methods, the insufficient number of available samples and inaccurate prior knowledge can cause serious deterioration of STAP performance. To tackle this problem, a new KA covariance matrix estimation algorithm based on geometric mean is proposed. Specifically, by utilizing the persymmetric structure information as well as the geometric mean in the space of positive definite matrices, a geometric mean estimator is constructed, which is then linearly combined with the prior covariance matrix to obtain a new type of CCM estimators. Simulation results illustrate that the proposed algorithm has good STAP performance with small samples and inaccurate prior knowledge. © The Institution of Engineering & technology 2023.

关键词： Covariance matrix

来源：评论

学校读者我要写书评

暂无评论

Speaker Recognition Based on Pre-Trained Model and Deep Clustering

Speaker Recognition Based on Pre-Trained Model and Deep Clus...

引用

IEEE International Conference on Multimedia and Expo (ICME)

作者： Liang He Zhida Song Shuanghong Liu Mengqi Niu Ying Hu Hao Huang School of Computer Science and Technology Xinjiang University Urumqi China School of Intelligence Science and Technology Xinjiang University Urumqi China Xinjiang Key Laboratory of Signal Detection and Processing Urumqi China Department of Electronic Engineering Tsinghua University Beijing China

ISBN: (数字)9798350390155

ISBN: (纸本)9798350390162

In this paper, we propose a novel loss by integrating a deep clustering (DC) loss at the frame-level and a speaker recognition loss at the segment-level into a single network without additional data requirements and exhaustive computation. The DC loss implicitly generates soft pseudo-phoneme labels for each frame-level feature, which facilitates extracting more discriminant speaker representation by suppressing phonetic content information. We study the DC loss not only on the acoustic feature, but also on the features extracted by the pre-trained models, such as wav2vec 2.0, HuBERT and WavLM. Experimental results on the VoxCeleb dataset shows that the overall system performance based on the pre-trained model features are better than the one on the acoustic feature. The proposed loss is significantly effective for systems on the acoustic feature and has a marginal improvement for systems on the pre-trained model feature.

关键词： Training Costs Computational modeling System performance Phonetics Multimedia databases Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Digital image biological detection technology based on the porous silicon periodic crystals film

引用

Optoelectronics Letters 2021年第9期17卷 552-557页

作者： YANG Jianfeng JIA Zhenhong LüXiaoyi HUANG Xiaohui WANG Jiajia School of Physical Science and Technology Xinjiang UniversityUrumqi 830046China School of Information Science and Engineering Xinjiang UniversityUrumqi 830046China Key Laboratory of Signal Detection and Processing Xinjiang Uygur Autonomous RegionXinjiang UniversityUrumqi 830046China

In order to solve this problem,a new biological detection method is proposed,which makes use of the characteristics of optical transmission at the edge of the spectral band gap and sensitive to refractive index *** the probe light with wavelength at the edge of the Bragg band gap of porous silicon is incident on the surface of porous silicon,the change of refractive index caused by deoxyribonucleic acid(DNA)reaction in porous silicon will affect the detection light intensity transmitted from the porous silicon *** analyzing the change of the average gray value of the transmitted light image,the concentration of the DNA can be obtained.

关键词： porous image refractive

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：