检索结果-内蒙古大学图书馆

2022 IEEE International Conference on Multimedia and Expo, ICME 2022

作者： Xiang, Xijie Zhu, Lin Li, Jianing Tian, Yonghong Huang, Tiejun Shanghai Jiao Tong University Department of Electrical and Computer Engineering China School of Computer Science Peking University China Peng Cheng National Laboratory China

ISBN: (数字)9781665485630

ISBN: (纸本)9781665485630

The event camera is a novel bio-inspired vision sensor. When the brightness change exceeds the preset threshold, the sensor generates events asynchronously. The number of valid events directly affects the performance of event-based tasks, such as reconstruction, detection, and recognition. However, when in low-brightness or slow-moving scenes, events are often sparse and accompanied by noise, which poses challenges for event-based tasks. To solve these challenges, we propose an event temporal up-sampling algorithm 11Code: https://***/XIJIE-XIANG/Event-Temporal-Up-sampling to generate more effective and reliable events. The main idea of our algorithm is to generate up-sampling events on the event motion trajectory. First, we estimate the event motion trajectory by contrast maximization algorithm and then up-sampling the events by temporal point processes. Experimental results show that up-sampling events can provide more effective information and improve the performance of downstream tasks, such as improving the quality of reconstructed images and increasing the accuracy of object detection. © 2022 IEEE.

关键词： Cameras

来源：评论

学校读者我要写书评

暂无评论

Scaling Distributed Collaboration in Mixed Reality

Scaling Distributed Collaboration in Mixed Reality

引用

2025 CHI Conference on Human Factors in Computing systems, CHI EA 2025

作者： Genay, Adélaïde Syiem, Brandon Victor Wong, Emily Feuchtner, Tiare Knibbe, Jarrod Grønbæk, Jens Emil Sloth Velloso, Eduardo School of Computer Science University of Sydney SydneyNSW Australia School of Computing and Information Systems The University of Melbourne MelbourneVIC Australia HCI Group University of Konstanz Konstanz Germany Department of Computer Science Aarhus University Aarhus Denmark School of Electrical Engineering and Computer Science The University of Queensland St. LuciaQLD Australia

ISBN: (纸本)9798400713958

Distributed collaboration in Mixed Reality (MR) promises to revolutionise how people connect across different physical environments, offering experiences akin to face-to-face interactions. However, previous work has mostly focused on enabling this vision in overly simplified settings such as with only two users interacting in identical distributed environments. Scaling current systems to work with large groups and for common real-life scenarios is a persistent challenge that requires addressing multiple tensions. We identified six challenges: 1) supporting locally congruent actions from heterogeneous remote spaces, 2) communicating accurate user behaviours through virtual representation instead of physical bodies, 3) facilitating organic group interactions within limited physical space, 4) maintaining conversational dynamics even in asynchronous exchanges, 5) providing equal access to physical objects for all participants, and 6) enabling efficient task switching within a complex ecology of applications, devices, and accessibility needs. This workshop aims to gather researchers and practitioners to explore actionable strategies for resolving these challenges. Through a mix of presentations, hands-on activities, and group discussions, participants will generate new ideas and develop a research agenda to articulate the future of MR collaboration systems. The workshop outcomes will include a list of concrete next steps for the community to bring distributed MR collaboration at scale. © 2025 Copyright held by the owner/author(s).

关键词： Grid computing

来源：评论

学校读者我要写书评

暂无评论

Classifying Bicycle Infrastructure Using On-Bike Street-Level Images

Classifying Bicycle Infrastructure Using On-Bike Street-Leve...

引用

International Conference on Intelligent Transportation

作者： Kal Backman Ben Beck Dana Kulić Department of Electrical and Computer Systems Engineering School of Public Health and Preventive Medicine Monash University

ISBN: (数字)9798331505929

ISBN: (纸本)9798331505936

While cycling offers an attractive option for sus-tainable transportation, many potential cyclists are discouraged from taking up cycling due to the lack of suitable and safe infrastructure. Efficiently mapping cycling infrastructure across entire cities is necessary to advance our understanding of how to provide connected networks of high-quality infrastructure. Therefore we propose a system capable of classifying available cycling infrastructure from on-bike smartphone camera data. The system receives an image sequence as input, temporally analyzing the sequence to account for sparsity of signage. The model outputs cycling infrastructure class labels defined by a hi-erarchical classification system. Data is collected via participant cyclists covering 7,006Km across the Greater Melbourne region that is automatically labeled via a GPS and OpenStreetMap database matching algorithm. The proposed model achieved an accuracy of 95.38%, an increase in performance of 7.55% compared to the non-temporal model. The model demonstrated robustness to extreme absence of image features where the model lost only 6.6% in accuracy after 90% of images being replaced with blank images. This work is the first to classify cycling infrastructure using only street-level imagery collected from bike-mounted mobile phone cameras, while demonstrating robustness to feature sparsity via long temporal sequence analysis.

关键词： Accuracy Databases Bicycles Feature extraction Cameras Robustness Vectors Image sequences Data mining Global Positioning System

来源：评论

学校读者我要写书评

暂无评论

Impact of Gale Weather Events on Wind Power Generation 1

Impact of Gale Weather Events on Wind Power Generation

引用

2024 1st International Conference on Smart Grids and Power systems, ICSGPS 2024

作者： Zhang, Yuanhe Yang, Ming Wang, Bo Li, Menglin Yu, Yixiao School of Computer Science and Engineering Shandong University Shandong province Jinan China State Key Laboratory of Operation and Control of Renewable Energy & Storage Systems China Electric Power Research Institute Beijing China

Due to the influence of global warming, extreme wind weather occurs frequently, especially in extreme weather such as typhoons and cold waves, problems such as wind turbine shutdown, cutting out, and sudden changes in power generation power will occur, which brings great challenges to wind power forecast and safe and stable operation of the power grid. At present, the research and analysis of strong wind weather mostly stay at the single fan level, and the causes of abnormal changes in wind power output are mostly analyzed from the physical level. However, with the large-scale access of wind power to the power grid, the impact of extreme weather on power grid operation is becoming more and more significant, and it is particularly important to study the impact of strong wind weather events on wind power cluster output. This paper summarizes various types of wind power generation weather events, then screens key meteorological factors affecting power generation under different wind weather conditions based on Gaussian search and grey correlation weighting, and finally analyzes the meteorological impact on the actual operation of wind farm stations and clusters under typhoons and cold waves respectively, aiming at the problem of huge deviations in wind power forecast caused by wind. The next research direction and suggestions are put forward. © 2024 Published under licence by IOP Publishing Ltd.

关键词： Wind power

来源：评论

学校读者我要写书评

暂无评论

AILS-NTUA at SemEval-2024 Task 6: Efficient model tuning for hallucination detection and analysis

arXiv

引用

arXiv 2024年

作者： Griogoriadou, Natalia Lymperaiou, Maria Filandrianos, Giorgos Stamou, Giorgos School of Electrical and Computer Engineering AILS Laboratory National Technical University of Athens Greece

In this paper, we present our team's submissions for SemEval-2024 Task-6 - SHROOM, a Shared-task on Hallucinations and Related Observable Overgeneration Mistakes. The participants were asked to perform binary classification to identify cases of fluent overgeneration hallucinations. Our experimentation included fine-tuning a pre-trained model on hallucination detection and a Natural Language Inference (NLI) model. The most successful strategy involved creating an ensemble of these models, resulting in accuracy rates of 77.8% and 79.9% on model-agnostic and model-aware datasets respectively, outperforming the organizers' baseline and achieving notable results when contrasted with the top-performing results in the competition, which reported accuracies of 84.7% and 81.3% correspondingly. © 2024, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Interview with light-wave architect Jelena Vuckovic

引用

Advanced Photonics 2024年第4期6卷 9-13页

作者： Mengjie Yu Jelena Vuckovic University of Southern California Viterbi School of EngineeringDepartment of Electrical and Computer EngineeringLos AngelesCaliforniaUnited States Stanford University the National Academy of Sciences the Nanoscale and Quantum Photonics Lab in the Ginzton Laboratory

Mengjie Yu:Hello,*** interview is organized by the journal Advanced *** today we're really honored and privileged to have Professor Jelena Vuckovic from Stanford University here with us.I'm from University of ... 详细信息

关键词： wave interview light

来源：评论

学校读者我要写书评

暂无评论

Deep Learning in Remote Sensing for Climate-Induced Disaster Resilience: A Comprehensive Interdisciplinary Approach

引用

Remote Sensing in Earth systems Sciences 2024年第1期8卷 145-160页

作者： Padmaja, S.M. Naveenkumar, R. Kumari, N. P. Lavanya Pimo, Er. S. John Bindhu, M. Konduri, Bhagavan Jangir, Pradeep Department of Electrical and Electronics Engineering Shri Vishnu Engineering College for Women Bhimavaram India Department of Computer Science and Engineering Chandigarh Colleges of Engineering Chandigarh Group of Colleges Jhanjeri Punjab Mohali 140307 India Department of Computer Science and Systems Engineering AUCE Andhra University Visakhapatnam India Department of Computer Science and Engineering St. Xavier’s Catholic College of Engineering Nagercoil India Department of ECE Sri Venkateswara College of Engineering Sriperumbudur India Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Andhra Pradesh Vaddeswaram India Department of Biosciences Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Chennai India Hourani Center for Applied Scientific Research Al-Ahliyya Amman University Amman 19328 Jordan

Climate-induced disasters pose significant threats to human lives, infrastructure, and ecosystems. Deep learning techniques, combined with remote sensing, offer powerful tools for predicting, detecting, and mitigating the impacts of such disasters. This study presents a comprehensive interdisciplinary approach using a hybrid cascaded CNN-RNN model to perform spatio-temporal analysis, enhanced image segmentation with U-Net and SegNet architectures, and a hybrid CNN-LSTM-XGBoost ensemble for multi-hazard prediction. The proposed CNN-RNN model achieves an accuracy of 96.78%, significantly outperforming traditional models in predicting flood-prone areas, wildfires, and droughts. Key metrics such as Intersection over Union (IoU), precision, recall, and F1-scores are used to evaluate model performance, highlighting the model’s ability to accurately identify disaster-prone areas and improve disaster resilience. This research not only demonstrates the efficacy of deep learning for climate-induced disaster prediction but also paves the way for future integration of socio-economic data to refine disaster risk assessment. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024.

关键词： Climate-induced disasters CNN-RNN Deep learning Disaster resilience Hybrid ensemble Multi-hazard prediction Remote sensing SegNet Spatio-temporal analysis U-Net

来源：评论

学校读者我要写书评

暂无评论

Towards Seamless Wireless Link Connection 23

Towards Seamless Wireless Link Connection

引用

21st Annual International Conference on Mobile systems, Applications and Services, MobiSys 2023

作者： Chen, Lili Yu, Bozhong Ren, Ju Gummeson, Jeremy Zhang, Yaoxue Department of Computer Science and Technology BNRist Tsinghua University China School of Information Science and Technology Northwest University United States Department of Electrical and Computer Engineering University of Massachusetts Amherst United States Zhongguancun Laboratory Beijing China

ISBN: (纸本)9798400701108

Sub-6GHz and mmWave complement each other in the next generation of wireless communications for wide coverage and high capacity. However, there is still a gap between current network technology and seamless connection in indoor environments due to the inevitable occlusions that particularly affect higher frequency bands like Wi-Fi 5GHz and mmWave. To overcome this gap, economical tunable metasurfaces offer a promising solution by redirecting the beam direction of incoming waves to bypass blockages. However, existing metasurface technologies focus on a single frequency band and lack a theoretical framework to guide surface design for multiple desired bands, limiting their potential for diverse *** work attempts to provide a general method for cross-band metasurface design, extend wireless link coverage by using a new metasurface prototype called CrossFlit, which can maximize the link quality with efficient aperture usage and low cost. We addressed a series of challenges to enable CrossFlit to independently control the wireless communication links across multiple frequency bands;our proof of concept implementation is for Wi-Fi 5 GHz and mmWave bands. We integrate a fast beam-scanning algorithm into an end-to-end system that actuates the metasurface to optimize links in realtime. Empirical evaluation demonstrates that CrossFlit can enhance multiple wireless links simultaneously, providing 10 dB and 15 dB power improvements for Wi-Fi and mmWave links, respectively. CrossFlit achieves less than 4° beam mismatch even in challenging mobile scenarios. © 2023 Owner/Author(s)

关键词： Wi-Fi

来源：评论

学校读者我要写书评

暂无评论

The Role of Predictive Maintenance Optimization Techniques in Enhancing Industrial Productivity 10

The Role of Predictive Maintenance Optimization Techniques i...

引用

10th International Conference on Optimization and Applications, ICOA 2024

作者： Er-Ratby, Mohamed Kobi, Abdessamad Sadraoui, Youssef Kadiri, Moulay Saddik National School of Applied Sciences Sultan Moulay Slimane University LASTI Laboratory of Science and Engineering Techniques Khouribga Morocco Polytech Angers University of Angers LARIS Angevin Laboratory for Systems Engineering Research Angers France National School of Applied Sciences Sultan Moulay Slimane University LIPIM Laboratory of Process Engineering Computer Science and Mathematics Khouribga Morocco

ISBN: (纸本)9798350387353

The convergence between advanced technology and industrial production holds significant potential to enhance operational efficiency, reduce maintenance costs, and minimize unexpected disruptions. Artificial intelligence, in particular, plays a crucial role in predicting machine failures, thereby contributing to the optimization of operational performance and maximizing overall company output. However, current challenges in maintenance and production are numerous. High maintenance costs are often a major concern for companies, as they can significantly eat into profit margins. Unexpected downtime, caused by sudden machine failures, leads to substantial production losses, affecting not only revenue but also customer satisfaction and the company's reputation. Furthermore, the increasing complexity of industrial systems makes the management of maintenance and production even more difficult, necessitating more advanced and integrated solutions. This article explores the various applications of artificial intelligence in the manufacturing industry, highlighting its impact on business performance. It also examines the steps necessary for implementing predictive maintenance, from data collection and processing to the establishment of predictive models. The ultimate goal is to improve equipment availability, optimize operational performance, and maximize overall company output. © 2024 IEEE.

关键词： Predictive maintenance

来源：评论

学校读者我要写书评

暂无评论

The group robustness is in the details: revisiting finetuning under spurious correlations 24

The group robustness is in the details: revisiting finetunin...

引用

Proceedings of the 38th International Conference on Neural Information Processing systems

作者： Tyler LaBonte John C. Hill Xinchen Zhang Vidya Muthukumar Abhishek Kumar H. Milton Stewart School of Industrial and Systems Engineering Georgia Institute of Technology School of Electrical and Computer Engineering Georgia Institute of Technology School of Electrical and Computer Engineering Georgia Institute of Technology and H. Milton Stewart School of Industrial and Systems Engineering Georgia Institute of Technology Google DeepMind

ISBN: (纸本)9798331314385

Modern machine learning models are prone to over-reliance on spurious correlations, which can often lead to poor performance on minority groups. In this paper, we identify surprising and nuanced behavior of finetuned models on worst-group accuracy via comprehensive experiments on four well-established benchmarks across vision and language tasks. We first show that the commonly used class-balancing techniques of mini-batch upsampling and loss upweighting can induce a decrease in worst-group accuracy (WGA) with training epochs, leading to performance no better than without class-balancing. While in some scenarios, removing data to create a class-balanced subset is more effective, we show this depends on group structure and propose a mixture method which can outperform both techniques. Next, we show that scaling pretrained models is generally beneficial for worst-group accuracy, but only in conjunction with appropriate class-balancing. Finally, we identify spectral imbalance in finetuning features as a potential source of group disparities — minority group covariance matrices incur a larger spectral norm than majority groups once conditioned on the classes. Our results show more nuanced interactions of modern finetuned models with group robustness than was previously known. Our code is available at https://***/tmlabonte/revisiting-finetuning.

关键词：

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：