检索结果-内蒙古大学图书馆

Region Aware Video Object Segmentation with Deep Motion Modeling

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Miao, Bo Bennamoun, Mohammed Gao, Yongsheng Mian, Ajmal The Department of Computer Science and Software Engineering The University of Western Australia Crawley PerthWA6009 Australia The School of Engineering Griffith University BrisbaneQLD4111 Australia

Current semi-supervised video object segmentation (VOS) methods usually leverage the entire features of one frame to predict object masks and update memory. This introduces significant redundant computations. To reduce redundancy, we present a Region Aware Video Object Segmentation (RAVOS) approach that predicts regions of interest (ROIs) for efficient object segmentation and memory storage. RAVOS includes a fast object motion tracker to predict their ROIs in the next frame. For efficient segmentation, object features are extracted according to the ROIs, and an object decoder is designed for object-level segmentation. For efficient memory storage, we propose motion path memory to filter out redundant context by memorizing the features within the motion path of objects between two frames. Besides RAVOS, we also propose a large-scale dataset, dubbed OVOS, to benchmark the performance of VOS models under occlusions. Evaluation on DAVIS and YouTube-VOS benchmarks and our new OVOS dataset show that our method achieves state-of-the-art performance with significantly faster inference time, e.g., 86.1 J&F at 42 FPS on DAVIS and 84.4 J&F at 23 FPS on YouTube-VOS. © 2022, CC BY-NC-SA.

关键词： Large dataset

Optimized Flame Detection Using a YOLOv5 Neural Network

学校读者我要写书评

暂无评论

Optimized Flame Detection Using a YOLOv5 Neural Network

Computing, Robotics and System sciences (ICRSS), International Conference on

作者： Shahid Karim Muhammad Ibrar Anas Bilal Akeel Qadir Irfana Bibi School of Information Engineering Xi'an Eurasia University Xi'an China Software College Shenyang Normal University Shenyang China College of Information Science and Technology Hainan Normal University Haikou China Department of Computer Science Faculty of Computing and Information Technology University of Punjab Lahore Punjab Pakistan

ISBN: (数字)9798331511296

ISBN: (纸本)9798331511302

With the rising prevalence of smart homes, there's an increasing demand for comprehensive automation solutions to mitigate fire risks, especially when homeowners are absent or in homes with elderly residents. This study focuses on enhancing flame detection techniques, particularly for smaller, hard-to-detect fire sources. A customized Yolov5 neural network was developed to improve detection precision. This enhancement was facilitated using a dataset of over 2,000 images from lab camera observations and online platforms. These images were then manually labeled and clustered for anchor dimensions using the K-means++ algorithm. The CIOU and diou _ nms techniques were integrated to optimize the network's performance. Upon training, the refined model was tested on the Jet Nano, revealing its high efficacy in flame detection.

关键词： YOLO Training Accuracy Sensitivity Service robots Neural networks Fires Clustering algorithms Smart homes Safety

Diegif: An Efficient and Secured Dicom to Egif Conversion Framework for Confidentiality in Machine Learning Training

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

作者： Siam, Abdullah Al Hassan, Md Maruf Rahaman, Md Atikur Abdullah, Masuk Department of Software Engineering Daffodil International University Daffodil Smart City Birulia Dhaka Dhaka1216 Bangladesh Department of Computer Science & Engineering Southeast University Tejgaon Dhaka1208 Bangladesh School of Economics and Management Jiujiang University 551 Qianjin Donglu Jiangxi Jiujiang332005 China Department of Vehicles Engineering Faculty of Engineering University of Debrecen Ótemető street. 2-4 Debrecen4028 Hungary

Medical imaging plays a critical role in contemporary healthcare, although it confronts issues relating to storage, security, and confidentiality in machine learning-based diagnostic systems. The proposed framework, Diegif, presents an efficient and safe mechanism for converting DICOM (Digital Imaging and Communications in Medicine) data into EGIF (Encrypted Graphics Interchange Format) files to overcome these challenges. The framework comprises four key components: (1) converting DICOM files to GIF format with encryption, (2) decrypting EGIF files for processing, (3) enabling confidentiality preserving machine learning training using EGIF data, and (4) facilitating physician diagnosis and report generation based on trained machine learning models. The Diegif framework aims to enhance storage efficiency by decreasing file sizes by 66.32%, thereby improving data transport efficacy and cloud storage affordability while preserving strong encryption for data confidentiality. Pseudocode algorithms are provided for each phase, ensuring reproducibility and transparency. This paper illustrates the Diegif framework’s potential to medical image processing, secure storage, and AI-driven diagnostic functions in healthcare. © 2024, The Authors. All rights reserved.

关键词： Medical image processing

The impact of vaccination on the infection rate and the severity of Covid-19

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Rayhan, Md. Abdullah-Al-Wadud, M. Ahmed, M. Helal Uddin Department Of Computer Science And Engineering Brac University Dhaka Bangladesh Department Of Software Engineering College Of Computer And Information Sciences King Saud University Riyadh Saudi Arabia Department Of Management Information Systems University Of Dhaka Dhaka1000 Bangladesh

This study aims to statistically assess the effectiveness of vaccination against SARS-CoV-2. It is indispensable to investigate the relationship between Covid-19 deadliness and vaccination in order to study the impact of vaccine in real-world. We studied rates of infection and death due to Covid-19 in different countries with respect to their levels of vaccination. People who received the required dose of vaccination were considered as fully vaccinated in this study. Based on the percentage of fully vaccinated population, countries were categorized into several groups. Though a high-level study on the vaccine effectiveness may not provide much insight for individual level differences, a global analysis is imperative to infer the influence of vaccination as a controlling measure of the pandemic. © 2022, CC BY-NC-ND.

关键词： Vaccines

Multi-objective Optimization for Multi-UAV-assisted Mobile Edge Computing

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Sun, Geng Wang, Yixian Sun, Zemin Wu, Qingqing Kang, Jiawen Niyato, Dusit Leung, Victor C.M. College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering Ministry of Education Jilin University Changchun130012 China College of Computing and Data Science Nanyang Technological University Singapore639798 Singapore Department of Electronic Engineering Shanghai Jiao Tong University Shanghai China School of Automation Guangdong University of Technology Guangzhou510006 China College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China Department of Electrical and Computer Engineering University of British Columbia VancouverBCV6T 1Z4 Canada

Recent developments in unmanned aerial vehicles (UAVs) and mobile edge computing (MEC) have provided users with flexible and resilient computing services. However, meeting the computing-intensive and latency-sensitive demands of users poses a significant challenge due to the limited resources of UAVs. To address this challenge, we present a multi-objective optimization approach for multi-UAV-assisted MEC systems. First, we formulate a multi-objective optimization problem aiming at minimizing the total task completion delay, reducing the total UAV energy consumption, and maximizing the total amount of offloaded tasks by jointly optimizing task offloading, computation resource allocation, and UAV trajectory control. Since the problem is a mixed-integer non-linear programming (MINLP) and NP-hard problem which is challenging, we propose a joint task offloading, computation resource allocation, and UAV trajectory control (JTORATC) approach to solve the problem. However, since the decision variables of task offloading, computation resource allocation, and UAV trajectory control are coupled with each other, the original problem is split into three sub-problems, i.e., task offloading, computation resource allocation, and UAV trajectory control, which are solved individually to obtain the corresponding decisions. Moreover, the sub-problem of task offloading is solved by using distributed splitting and threshold rounding methods, the sub-problem of computation resource allocation is solved by adopting the Karush-Kuhn-Tucker (KKT) method, and the sub-problem of UAV trajectory control is solved by employing the successive convex approximation (SCA) method. Simulation results show that the proposed JTORATC has superior performance compared to the other benchmark methods. Copyright © 2024, The Authors. All rights reserved.

关键词： Unmanned aerial vehicles (UAV)

A Lightweight Cloud-Edge Collaborative Intelligence Inference Framework with Runtime Dynamic Optimization for Resource-Constrained Consumer Electronics

学校读者我要写书评

暂无评论

IEEE Transactions on Consumer Electronics 2025年

作者： Wang, Chenlu Peng, Yuhuai Zhang, Dawei Alturki, Ryan Alshawi, Bandar Alotaibi, Majid Northeastern University College of Computer Science and Engineering Shenyang110819 China Zhiyuan Research Institute Xihu District Hangzhou China Zhengzhou University School of Computer and Artificial Intelligence Henan Zhengzhou450001 China Umm Al-Qura University College of Computing Department of Software Engineering Makkah24382 Saudi Arabia Umm Al-Qura University College of Computing Department of Computer and Network Engineering Makkah24382 Saudi Arabia

The proliferation of Internet of Things (IoT) and embedded computing has led to widespread deployment of smart consumer electronics requiring edge-based Artificial Intelligence (AI) capabilities. However, the heterogeneous nature of sensing data and dynamic edge environments poses significant challenges for efficient model inference on resource-constrained devices. To address these challenges, this paper presents a lightweight collaborative inference framework designed for consumer electronics. First, we formulate the inference optimization problem as a Mixed-Integer Nonlinear Programming (MINLP) problem, considering channel pruning, early exit and cloud offloading decisions to optimize the trade-off between accuracy and computational cost. Second, we propose a selective model activation mechanism based on Markov Decision Process (MDP), which employs a recursive self-attention mechanism to dynamically track inference budgets and guide decision-making through encoder-decoder architectures. The mechanism integrates entropy regularization during training to ensure robust and diverse execution paths. Comprehensive experiments demonstrate that our framework achieves 65.50% reduction in model parameters and 80.68% reduction in inference Floating Point Operations (FLOPs) while maintaining accuracy loss within 0.81% of the original model, making it suitable for real-time AI applications on resource-constrained consumer electronics. © 1975-2011 IEEE.

关键词： Markov processes

Mobile Robot Motion Stability and Optimal Chassi Construction 11

学校读者我要写书评

暂无评论

Mobile Robot Motion Stability and Optimal Chassi Constructio...

11th International Conference on Advanced computer Information Technologies, ACIT 2021

作者： Kuzmych, Olena Cherniashchuk, Nataliia Lishchyna, Nataliia Lishchyna, Valeriy Mekush, Oksana Gumenyuk, Pavlo Lutsk National Technical University Department of Computer Engineering and Cyber Security Lutsk Ukraine Lutsk National Technical University Department of Software Engineering Lutsk Ukraine Lutsk National Technical University Department of Computer Science Lutsk Ukraine Eastern European National University Department of Applied Mathematics Lutsk Ukraine Lutsk National Technical University Department of Automation and Computer Integrated Technologies Lutsk Ukraine

The paper presents the results of dynamical modeling, simulation and stability analysis of mobile robot, which makes it possible to explain some dynamic effects that lead to instability of the straight-line robot's movement and skidding at bends at high speeds. This model is used to analyze the stable motion of wheeled robot in accordance to mutual placement of the wheels and the center of mass during its movement along the straight line. The appropriate recommendations are given on the optimal chassis design to avoid the phenomenon of instability motion. © 2021 IEEE.

关键词： Mobile robots

UX Design and Evaluation: Holism versus Reductionism Approaches - Which is Better?

学校读者我要写书评

暂无评论

UX Design and Evaluation: Holism versus Reductionism Approac...

Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI), IEEE Conference on

作者： Emmanuel O.C. Mkpojiogu Monday Jubrin Abdullahi Obianuju E. Okeke-Uzodike Universiti Utara Malaysia 06010 UUM Sintok Malaysia Department of Software Engineering Veritas University Abuja Nigeria Department of Computer Science Airforce Institute of Technology Kaduna Kaduna State Nigeria West Africa Department of Applied Management Durban University of Technology Pietermaritzburg KwaZulu-Natal South Africa

A literature content analysis strategy was used to synthesize extant literature sources to comprehend and capture the approaches to the design, evaluation and modeling of user experience. From the analysis, two main methods were identified for the UX research. These paradigms are holism and reductionism. The holist approach stresses on sense making and interpretative approaches to user experience, leading to construction and reconstruction of users’ quality perceptions about their experiences with technological artifacts. This paradigm is design based. Conversely, the reductionism approach is concerned with the deconstruction of users’ experiences. This approach conceives experiences as a model that can be broken down into aspects and dimensions that facilitates understanding of user experience. The debate among researchers on these differing viewpoints conceived and drove this study. The findings of the research show that the two paradigms are insightful, rich and unique and can be independently used in UX research, nevertheless, as both approaches a limited and address the peculiar concerns of the two respective paradigms, this paper advocates and endorses the use of a mixed paradigm especially in longitudinal studies to amalgamate the rich components of both varying approaches, thus implying the interdependent use of the two paradigms in UX research.

关键词：

Vec2Face-v2: Unveil Human Faces from their Blackbox Features via Attention-based Network in Face Recognition

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Truong, Thanh-Dat Duong, Chi Nhan Le, Ngan Savvides, Marios Luu, Khoa The Computer Vision and Image Understanding Lab University of Arkansas United States The Department of Computer Science and Software Engineering Concordia University Canada Carnegie Mellon University United States

In this work, we investigate the problem of face reconstruction given a facial feature representation extracted from a blackbox face recognition engine. Indeed, it is a very challenging problem in practice due to the limitations of abstracted information from the engine. We, therefore, introduce a new method named Attention-based Bijective Generative Adversarial Networks in a Distillation framework (DAB-GAN) to synthesize the faces of a subject given his/her extracted face recognition features. Given any unconstrained unseen facial features of a subject, the DAB-GAN can reconstruct his/her facial images in high definition. The DAB-GAN method includes a novel attention-based generative structure with the newly defined Bijective Metrics Learning approach. The framework starts by introducing a bijective metric so that the distance measurement and metric learning process can be directly adopted in the image domain for an image reconstruction task. The information from the blackbox face recognition engine will be optimally exploited using the global distillation process. Then an attention-based generator is presented for a highly robust generator to synthesize realistic faces with ID preservation. We have evaluated our method on the challenging face recognition databases, i.e., CelebA, LFW, CFP-FP, CP-LFW, AgeDB, CA-LFW, and consistently achieved state-of-the-art results. The advancement of DAB-GAN is also proven in both image realism and ID preservation properties. Copyright © 2022, The Authors. All rights reserved.

关键词： Face recognition