检索结果-内蒙古大学图书馆

DCSF-KD: Dynamic Channel-wise Spatial Feature Knowledge Distillation for Object Detection 39

学校读者我要写书评

暂无评论

DCSF-KD: Dynamic Channel-wise Spatial Feature Knowledge Dist...

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Dai, Tao Lin, Yang Guo, Hang Wang, Jinbao Zhu, Zexuan College of Computer Science and Software Engineering Shenzhen University China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China Tsinghua Shenzhen International Graduate School Tsinghua University Shenzhen China Shenzhen City Key Laboratory of Embedded System Design Shenzhen China Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen China

ISBN: (纸本)157735897X

Knowledge distillation (KD) has recently gained great success in the field of object detection. By transferring the knowledge of the spatial or channel domain from the teacher model to the student model, it allows for a more compact representation with minimal performance loss. Despite this progress, existing KD methods typically treat knowledge from spatial or channel domains independently, ignoring the exploitation of the mutual relationship between these domains. In this work, we first explore the connection between spatial and channel domains and find there exists a strong correlation between them, i.e. the salient channels tend to contain significant object regions in the spatial domain. Motivated by this observation, we propose DCSF-KD, a novel Dynamic Channel-wise Spatial Feature Knowledge Distillation framework for object detection by fully exploiting both spatial and channel knowledge. Specifically, we introduce channel-wise spatial feature distillation and global channel attention distillation, using information from both domains to improve the accuracy of the student network. Experiments demonstrate that our DCSF-KD outperforms existing detection methods on both homogeneous and heterogeneous teacher-student network pairs. For example, when using the MaskRCNN-Swin detector as the teacher, and based on RetinaNet and FCOS with ResNet-50 on MS COCO, our DCSF-KD can achieve 41.9% and 44.1% mAP, respectively. © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Students

Dual Perceptual Loss for Single Image Super-Resolution Using ESRGAN

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Song, Jie Yi, Huawei Xu, Wenqian Li, Xiaohui Li, Bo Liu, Yuanyuan School of Electronics and Information Engineering Liaoning University of Technology Jinzhou121001 China Key Laboratory of Advanced Design and Intelligent Computing Ministry of Education Dalian University Dalian116622 China

The proposal of perceptual loss solves the problem that per-pixel difference loss function causes the reconstructed image to be overly-smooth, which acquires a significant progress in the field of single image super-resolution reconstruction. Furthermore, the generative adversarial networks (GAN) is applied to the super-resolution field, which effectively improves the visual quality of the reconstructed image. However, under the condtion of high upscaling factors, the excessive abnormal reasoning of the network produces some distorted structures, so that there is a certain deviation between the reconstructed image and the ground-truth image. In order to fundamentally improve the quality of reconstructed images, this paper proposes a effective method called Dual Perceptual Loss (DP Loss), which is used to replace the original perceptual loss to solve the problem of single image super-resolution reconstruction. Due to the complementary property between the VGG features and the ResNet features, the proposed DP Loss considers the advantages of learning two features simultaneously, which significantly improves the reconstruction effect of images. The qualitative and quantitative analysis on benchmark datasets demonstrates the superiority of our proposed method over state-of-the-art super-resolution methods. Copyright © 2022, The Authors. All rights reserved.

关键词： Generative adversarial networks

LCFNet: Loss Compensation Fusion Network for Real-Time Semantic Segmentation of Urban Road Scenes

学校读者我要写书评

暂无评论

LCFNet: Loss Compensation Fusion Network for Real-Time Seman...

International Conference on Intelligent Transportation

作者： Lu Yang Yiwen Bai Fenglei Ren Shiyu Zhang Chongke Bi Tianjin Key Laboratory for Advanced Mechanical System Design and Intelligent Control School of Mechanical Engineering Tianjin University of Technology Tianjin China National Demonstration Center for Experimental Mechanical and Electrical Engineering Education (Tianjin University of Technology) Tianjin China College of Intelligence and Computing Tianjin University Tianjin China

Semantic segmentation is used by intelligent transportation systems to understand and sense the traffic environment. However, achieving semantic segmentation in real-time is a challenge due to the necessity of both high accuracy and fast processing. This is especially valuable for applications such as autonomous driving and industrial robotics. In this paper, we propose a real-time semantic segmentation network, called LCFNet, which makes use of three-branch structure. The LCFNet consists of Lightweight Detail Guidance Fusion (L-DGF) and Lightweight Semantic Guidance Fusion (L-SGF) modules. Both modules aggregate information from various network layers. In the termination of network, a Total Guidance Fusion (TGF) module is proposed for processing information from all three branches. Depth-wise Convolution Pyramid Pooling (DCPP) module is also included to optimize accuracy and simplify computation. The effectiveness of LCFNet is demonstrated on two typical semantic segmentation datasets, Cityscapes and CamVid. On a single NVIDIA GeForce GTX 2080 Ti GPU, LCFNet reaches 77.02% mIoU at 95.97 FPS and 81.17% mIoU at 204.82 FPS, respectively.

关键词：

Generative AI for Data Augmentation in Wireless Networks: Analysis, Applications, and Case Study

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wen, Jinbo Kang, Jiawen Niyato, Dusit Zhang, Yang Wang, Jiacheng Sikdar, Biplab Zhang, Ping College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics China School of Automation Guangdong University of Technology China College of Computing and Data Science Nanyang Technological University Singapore Department of Electrical and Computer Engineering College of Design and Engineering National University of Singapore Singapore State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications China

Data augmentation is a powerful technique to mitigate data scarcity. However, owing to fundamental differences in wireless data structures, traditional data augmentation techniques may not be suitable for wireless data. Fortunately, Generative Artificial Intelligence (GenAI) can be an effective alternative to wireless data augmentation due to its excellent data generation capability. This article systemically explores the potential and effectiveness of GenAI-driven data augmentation in wireless networks. We first briefly review data augmentation techniques, discuss their limitations in wireless networks, and introduce generative data augmentation, including reviewing GenAI models and their applications in data augmentation. We then explore the application prospects of GenAI-driven data augmentation in wireless networks from the physical, network, and application layers, which provides a GenAI-driven data augmentation architecture for each application. Subsequently, we propose a general generative diffusion model-based data augmentation framework for Wi-Fi gesture recognition, which uses transformer-based diffusion models to generate high-quality channel state information data. Furthermore, we develop residual neural network models for Wi-Fi gesture recognition to evaluate the role of augmented data and conduct a case study based on a real dataset. Simulation results demonstrate the effectiveness of the proposed framework. Finally, we discuss research directions for generative data augmentation. Copyright © 2024, The Authors. All rights reserved.

关键词： Generative adversarial networks

Enhancement of Co-located Shared VR Experiences: Representing Non-HMD Observers on Both HMD and 2D Screens

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Guo, Zixuan Xu, Wenge Wang, Hongyu Wan, Tingjie Baghaei, Nilufar Lo, Cheng-Hung Liang, Hai-Ning School of Advanced Technology Xi’an Jiaotong-Liverpool University Suzhou China Department of Computer Science The University of Liverpool Liverpool United Kingdom College of Computing Birmingham City University Birmingham United Kingdom School of Electrical Engineering and Computer Science The University of Queensland Brisbane Australia Department of Industrial Design Xi’an Jiaotong-Liverpool University Suzhou China Guangzhou China

Virtual reality (VR) not only allows head-mounted display (HMD) users to immerse themselves in virtual worlds but also to share them with others. When designed correctly, this shared experience can be enjoyable. However, in typical scenarios, HMD users are isolated by their devices, and non-HMD observers lack connection with the virtual world. To address this, our research investigates visually representing observers on both HMD and 2D screens to enhance shared experiences. The study, including five representation conditions, reveals that incorporating observer representation positively impacts both HMD users and observers. For how to design and represent them, our work shows that HMD users prefer methods displaying real-world visuals, while observers exhibit diverse preferences regarding being represented with real or virtual images. We provide design guidelines tailored to both displays, offering valuable insights to enhance co-located shared VR experiences for HMD users and non-HMD observers. © 2024, CC BY-NC-SA.

关键词： Virtual environments

Statistical Analysis of Digital Financial Technology Adoption Research

学校读者我要写书评

暂无评论

Statistical Analysis of Digital Financial Technology Adoptio...

Science, engineering and Business for Sustainable Development Goals (SEB-SDG), International Conference on

作者： Temiloluwa Iyanuoluwa Ajibade Olamide Emmanuel Ayodele Samuel-Soma M. Ajibade Nanet A. Goles Johnry P. Dayupay Anthonia Oluwatosin Adediran Department of Business Management Scholar School System Birmingham United Kingdom Department of Banking and Finance Ekiti State University Ado Ekiti Ekiti State Department of Computing and Information Systems School of Engineering and Technology Sunway University Selangor Malaysia Cebu Technological University Moalboal Cebu Philippines Faculty of Architecture and Urban Design Federal University of Uberlandia Uberlandia Brazil

ISBN: (数字)9798350358155

ISBN: (纸本)9798350358162

Digital financial technology adoption (DFTA) studies published in Scopus between 2019 and 2022 were examined in this study. After that, we assessed relevant literature, did bibliometric studies, and examined publication and citation patterns. The statistics showed 2643 citations and 176 publications. DFTA research has gained popularity when time-varying TP and TC values increased gradually. This topic's most popular sources are "articles" and "Sustainability (Switzerland)". A search of DFTA research stakeholders revealed several authors, organizations, and nations. Most famously associated with the University of Johannesburg in South Africa, David Mhlanga has produced more publications than anyone else. The National Natural Science Foundation and other Chinese donors who dominate DFTA research funding contributed to researchers' and affiliations' output. Cluster and hotspot analysis revealed two themes, "Digital financial technology adoption" and "Sustainable Banking Inclusion," and three keywords, Digital financial technology adoption, monetary inclusion, and finance. According to the examined literature, DFTA's effects on international financial stability and GDP growth are understudied. Future research should examine DFTA's environmental impact, long-term profitability, and commercial strategies for low-income and rural communities.

关键词： Productivity Statistical analysis Profitability Bibliometrics Banking Stability analysis Stakeholders

Application of Extreme Learning Machine for Shunt Faults Detection and Classification in Three-phase Transmission Line Systems

学校读者我要写书评

暂无评论

Application of Extreme Learning Machine for Shunt Faults Det...

Signal, Control and Communication (SCC), International Conference on

作者： Khaoula Assadi Jihane Ben Slimane Salah Salhi University of Tunis El Manar National Engineering School of Tunis LR11ES20 Analysis Design and Control of Systems Laboratory Tunis Tunisia Department of Computer Sciences Faculty of Computing and InformationTechnology Northern Border University Rafha Saudi Arabia

ISBN: (数字)9798350326390

ISBN: (纸本)9798350326406

The electrical power transmission lines play a crucial role in maintaining a continuous electricity supply. However, the exposed environment of these lines increases the risks of faults occurrence, necessitating prompt detection, and classification, to eliminate them within a specified time. In light of this, the present study introduces a methodology for the detection and classification of faults in a three-phase transmission line system employing the Extreme Learning Machine (ELM). A 220 kV, 300 km three-phase transmission line system was simulated in MATLAB/Simulink software, where the current and voltage signals were generated under different fault conditions including fault types, fault inception time, and fault resistance. The fundamental components of these signals were used as inputs in the ELM model. Different shunt fault types, including line-to-ground, line-to-line, double-line-to-ground, triple-line, and triple-line-to-ground were applied for fault detection and classification framework. Furthermore, the optimum number of hidden neurons in the ELM model was investigated in this study. In terms of the Mean Squared Error (MSE) and correlation coefficient, four ELM variants based on Sine, Sigmoid, Radial Basis, and Triangular Basis transfer functions were assessed for each module. The outcome of these metrics indicates that the ELM with the sigmoid model demonstrated superior performance compared to the other ELM models. Based on statistical metrics MSE, Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE), and also the computational time, the efficiency of ELM-Sigmoid and state-of-the-art methods of Multi-Layer Perceptron Neural Network (MLPNN) and Radial Basis Function Neural Network (RBFNN) were compared and analyzed during training and testing phases. The results obtained revealed that the ELM-Sigmoid model outperforms the other models in the detection and classification of shunt faults with significant accuracy and fast computational speed.

关键词： Measurement Training Power transmission lines Extreme learning machines Computational modeling Fault detection Neurons

Smart Traffic Management System Using RFID System

学校读者我要写书评

暂无评论

Smart Traffic Management System Using RFID System

MIT Art, design and Technology school of computing International Conference (MITADTSoCiCon)

作者： Shrikala Deshumkh Milind Gayakwad Nitin S. More Rohini Jadhav Kalyani Kadam Hanmant Magar Amity University Mumbai India College of Engineering Bharati Vidyapeeth (Deemed to be University) Pune India Department of CSE School of Computing MIT Art Design and Technology University Pune Symbiosis Institute of Technology Pune Symbiosis International (Deemed University) Pune India BRACT’s Vishwakarma Institute of Information Technology

ISBN: (数字)9798350362879

ISBN: (纸本)9798350362886

Traffic management is an essential aspect of administration. In the 21 st century since everything is going smart therefore our traffic management should also be smart. Congestion caused by heavy traffic is a major problem around the world. Accidents, increased trip time, poor law and order management, etc., are caused by improper management of the traffic. As the population is increasing day by day heavy traffic is becoming a greater problem. Due to these problems, the financial situations of states are heavily affected. Many so-called countries have smart management of their traffic, but this problem is greater in third-world countries or developing countries. In this paper, we are going to discuss a methodology that is used around the world to handle such problems and this method is the RFID system which is a radio frequency identification system.

关键词： Sociology Finance Developing countries Security Automobiles RFID tags Statistics

Wireless Hallucination in Generative AI-enabled Communications: Concepts, Issues, and Solutions

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Wang, Xudong Wang, Jiacheng Feng, Lei Niyato, Dusit Zhang, Ruichen Kang, Jiawen Xiong, Zehui Du, Hongyang Mao, Shiwen State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing100876 China College of Computing and Data Science Nanyang Technological University Singapore School of Automation Guangdong University of Technology Guangzhou510006 China Pillar of Information Systems Technology and Design Singapore University of Technology and Design Singapore Department of Electrical and Electronic Engineering University of Hong Kong Pok Fu Lam Hong Kong Department of Electrical and Computer Engineering Auburn University Auburn United States

Generative AI (GenAI) is driving the intelligence of wireless communications. Due to data limitations, random generation, and dynamic environments, GenAI may generate channel information or optimization strategies that violate physical laws or deviate from actual real-world requirements. We refer to this phenomenon as wireless hallucination, which results in invalid channel information, spectrum wastage, and low communication reliability but remains underexplored. To address this gap, this article provides a comprehensive concept of wireless hallucinations in GenAI-driven communications, focusing on hallucination mitigation. Specifically, we first introduce the fundamental, analyze its causes based on the GenAI workflow, and propose mitigation solutions at the data, model, and post-generation levels. Then, we systematically examines representative hallucination scenarios in GenAI-enabled communications and their corresponding solutions. Finally, we propose a novel integrated mitigation solution for GenAI-based channel estimation. At the data level, we establish a channel estimation hallucination dataset and employ generative adversarial networks (GANs)-based data augmentation. Additionally, we incorporate attention mechanisms and large language models (LLMs) to enhance both training and inference performance. Experimental results demonstrate that the proposed hybrid solutions reduce the normalized mean square error (NMSE) by 0.19, effectively reducing wireless hallucinations. Copyright © 2025, The Authors. All rights reserved.

关键词： Generative adversarial networks