Výsledky vyhľadávania - "edge devices"

Confounding factors-aware hierarchical deep learning for sustainable edge wildfire detection

Autori: El Madafri, Ismail

Prispievatelia: University/Department: Universitat Politècnica de Catalunya. Institut Universitari de Recerca en Ciència i Tecnologies de la Sostenibilitat

Thesis Advisors: Peña Carrera, Marta, Olmedo Torre, Noelia

Zdroj: TDX (Tesis Doctorals en Xarxa)

Predmety: wildfire detection, deep learning, confounding elements, multi-task learning, knowledge distillation, hierarchical learning, domain adaptation, edge devices, sustainable AI, environmental monitoring., detecció d'incendis forestals, aprenentatge profund, elements confusos, aprenentatge multitarea, destil·lació de coneixement, aprenentatge jeràrquic, adaptació al domini, dispositius edge, IA sostenible, monitoratge ambiental, detección de incendios forestales, aprendizaje profundo, elementos confusos, aprendizaje multitarea, destilación de conocimiento, aprendizaje jerárquico, adaptación al dominio, dispositivos edge, monitoreo ambiental., Àrees temàtiques de la UPC::Desenvolupament humà i sostenible, 502 - Natura. Estudi, conservació i protecció de la natura, 630 - Silvicultura. Arboricultura

Popis súboru: application/pdf

Prístupová URL adresa: http://hdl.handle.net/10803/695130
https://dx.doi.org/10.5821/dissertation-2117-440594

Optimizing TCN Inference: A Hardware-Software Co-Design Approach with CGRA Acceleration

Autori: Varaldi, Alessandro Naclerio, Alessio Riente, Fabrizio a ďalší

Zdroj: 2025 IEEE Computer Society Annual Symposium on VLSI (ISVLSI). :1-6

Predmety: Temporal Convolutional Network, Hardware-software Co-design, Deep Learning, Internet Of Things, Experimental Evaluation, Data Transfer, Inference System, Internet Of Things Applications, Edge Devices, Hardware Accelerators, Execution Efficiency, 1D Convolution, Internet Of Things Scenarios, Energy Efficiency, Convolutional Layers, Parallelization, Recurrent Neural Network, Convolution Operation, Dilated Convolution, Long-range Dependencies, Resource-constrained Environments, Efficient Inference, Intermediate Activity, Model Input Data, Memory Bank, Flash Memory, External Memory

Popis súboru: application/pdf

Prístupová URL adresa: https://ieeexplore.ieee.org/document/11130330/keywords#keywords
https://hdl.handle.net/11583/3002589
https://doi.org/10.1109/isvlsi65124.2025.11130330

Survey Based on Edge Structured File Systems in Edge Computing

Autori: Rajesh, Ravula Patgiri, Ripon

Zdroj: Tehnički Glasnik, Vol 19, Iss 2, Pp 305-312 (2025)
Tehnički glasnik
Volume 19
Issue 2

Predmety: Technology, edge environment, edge computing, cloud computing, edge devices, distributed file systems, edge structured file system, stakeholders, decentralized designs

Popis súboru: application/pdf

Prístupová URL adresa: https://doaj.org/article/83c31d4992844b9b924fa967e027d6f5
https://hrcak.srce.hr/329913
https://doi.org/10.31803/tg-20240111081254
https://hrcak.srce.hr/file/477071
https://hrcak.srce.hr/329913
https://doi.org/10.31803/tg-20240111081254

OLORAS: Online LOng Range Action Segmentation for Edge Devices

Autori: Ziche, Filippo Bombieri, Nicola

Zdroj: 2025 Design, Automation & Test in Europe Conference (DATE). :1-7

Predmety: temporal action segmentation, edge devices, linear recurrent units

A New Benchmark and Baseline for Real-Time High-Resolution Image Inpainting on Edge Devices

Autori: Sánchez-Ortega, Marcelo Triginer-Garces, Gil Ballester, Coloma a ďalší

Zdroj: 2025 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV). :1133-1143

Predmety: High-resolution inpainting, Real-time inpainting, Edge devices, Inpainting

Popis súboru: application/pdf

Prístupová URL adresa: http://hdl.handle.net/10230/69997

Learning-Enabled CPS for Edge-Cloud Computing

Autori: Capogrosso, Luigi Xu, Shengjie Fraccaroli, Enrico a ďalší

Zdroj: 2024 IEEE 14th International Symposium on Industrial Embedded Systems (SIES)

Predmety: Split Computing, Early Exit, Deep Neural Networks, Cyber-Physical Systems, Edge Devices, Split Computing, Early Exit, Deep Neural Networks, Cyber-Physical Systems, Edge Devices

Popis súboru: application/pdf

Dynamic Optimization of Recurrent Networks for Wind Speed Prediction on Edge Devices

Autori: Laeeq Aslam Runmin Zou Ebrahim Shahzad Awan a ďalší

Zdroj: IEEE Access, Vol 13, Pp 114520-114541 (2025)

Predmety: deep learning, Adaptive simulated annealing, edge devices, Electrical engineering. Electronics. Nuclear engineering, hyperparameter optimization, renewable energy, wind speed prediction, TK1-9971

Prístupová URL adresa: https://doaj.org/article/dd4ff121dea94c018ff35981b919e6ad

A Novel and Efficient Multi-Target Backdoor Attack for Deep Learning-Based Wireless Signal Classifiers

Autori: Nayan Moni Baishya B. R. Manoj Snehashish Karmakar

Zdroj: IEEE Access, Vol 13, Pp 65863-65883 (2025)

Predmety: backdoor attack, Activation clustering, deep learning, edge devices, Hadamard matrices, Electrical engineering. Electronics. Nuclear engineering, modulation classification, TK1-9971

Prístupová URL adresa: https://doaj.org/article/905b277171c74b2baef0c40ae9793f1e

Improving Performance of Real-Time Object Detection in Edge Device Through Concurrent Multi-Frame Processing

Autori: Seunghwan Kim Changjong Kim Sunggon Kim

Zdroj: IEEE Access, Vol 13, Pp 1522-1533 (2025)

Predmety: machine learning, Edge devices, object detection, Electrical engineering. Electronics. Nuclear engineering, performance optimization, TK1-9971

Prístupová URL adresa: https://doaj.org/article/2d4427bfc0f74dae851e3c9e2a5035d6

Ghost Module-Enhanced MTCNN: A Lightweight Cascade Framework for High-Accuracy Face Detection in Edge-Deployable Scenarios

Autori: Chen Wang Fen Liu

Zdroj: IEEE Access, Vol 13, Pp 107694-107709 (2025)

Predmety: multi-task cascaded convolutional networks, lightweight network, edge devices, Electrical engineering. Electronics. Nuclear engineering, Face detection, TK1-9971

Prístupová URL adresa: https://doaj.org/article/015718c66c224b9594db797498a7d589

StreamEase: Enabling Real-Time Inference of Temporal Convolution Networks on Low-Power MCUs with Stream-Oriented Automatic Transformation

Autori: Mirsalari, Seyed Ahmad Bijar, Léo Fariselli, Marco a ďalší

Zdroj: 2024 31st IEEE International Conference on Electronics, Circuits and Systems (ICECS)

Predmety: Temporal Convolutional Network Receptive Field Computational Demands Resource-constrained Devices Convolution Operation Network Output Per Cycle Input Size Accuracy Metrics Digital Signal Processing Number Of Time Steps Buffer Size Edge Devices Receptive Field Size Dilation Rate Multiple Time Steps Stream Model Memory Hierarchy

A Real-time Filter for Human Pose Estimation based on Denoising Diffusion Models for Edge Devices

Autori: Chiara Bozzini Michele Boldo Enrico Martini a ďalší

Zdroj: 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). :10864-10869

Predmety: Human pose estimation, Denoising Diffusion Models, Filtering, Edge Devices

Prístupová URL adresa: https://hdl.handle.net/11562/1135766

MTL-Split: Multi-Task Learning for Edge Devices using Split Computing

Autori: Luigi Capogrosso Enrico Fraccaroli Samarjit Chakraborty a ďalší

Zdroj: Proceedings of the 61st ACM/IEEE Design Automation Conference. :1-6

Predmety: FOS: Computer and information sciences, Computer Science - Machine Learning, Computer Science - Distributed, Parallel, and Cluster Computing, Computer Vision and Pattern Recognition (cs.CV), Split Computing, Multi-Task Learning, Deep Neural Networks, Edge Devices, Split Computing, Multi-Task Learning, Deep Neural Networks, Edge Devices, Computer Science - Computer Vision and Pattern Recognition, Distributed, Parallel, and Cluster Computing (cs.DC), Machine Learning (cs.LG)

Popis súboru: application/pdf

Prístupová URL adresa: http://arxiv.org/abs/2407.05982
https://dl.acm.org/doi/10.1145/3649329.3655686
https://doi.org/10.1145/3649329.3655686
https://hdl.handle.net/11562/1145092
https://dl.acm.org/doi/10.1145/3649329.3655686
https://doi.org/10.1145/3649329.3655686
https://hdl.handle.net/11583/2994382

Efficient attention vision transformers for monocular depth estimation on resource-limited hardware

Autori: Claudio Schiavella Lorenzo Cirillo Lorenzo Papa a ďalší

Zdroj: Scientific Reports, Vol 15, Iss 1, Pp 1-24 (2025)

Predmety: Computer vision, Edge devices, Efficient vision transformer, Monocular depth estimation, Medicine, Science

Popis súboru: electronic resource

Relation: https://doaj.org/toc/2045-2322

Prístupová URL adresa: https://doaj.org/article/3f3f6c565f8049b1801371c3507a4a40

Enhancing Split Computing and Early Exit Applications through Predefined Sparsity

Autori: Capogrosso, Luigi Fraccaroli, Enrico Petrozziello, Giulio a ďalší

Zdroj: 2024 Forum on Specification & Design Languages (FDL)

Predmety: FOS: Computer and information sciences, Deep Neural Networks, Computer Science - Machine Learning, Split Computing, Early Exit, Predefined Sparsity, Edge Devices, Computer Science - Distributed, Parallel, and Cluster Computing, Computer Vision and Pattern Recognition (cs.CV), Computer Science - Computer Vision and Pattern Recognition, Distributed, Parallel, and Cluster Computing (cs.DC), Machine Learning (cs.LG)

Popis súboru: application/pdf

Prístupová URL adresa: http://arxiv.org/abs/2407.11763
https://hdl.handle.net/11562/1137586
https://doi.org/10.1109/fdl63219.2024.10673767
https://ieeexplore.ieee.org/document/10673767

Imaging on the Edge: GPU-Accelerated Traveltime Tomography on a Jetson Nano

Autori: Shin, Ban-Sok Wientgens, Luis Shutin, Dmitriy

Zdroj: 2024 32nd European Signal Processing Conference (EUSIPCO). :601-605

Predmety: traveltime tomography, GPGPU, CUDA, edge devices, Seismic imaging

Prístupová URL adresa: https://elib.dlr.de/204379/
https://doi.org/10.23919/eusipco63174.2024.10715456

Deep Learning Based Speech Enhancement on Edge Devices Applied to Assistive Work Equipment

Autori: Nika Šljubura Marko Šimić Vedran Bilas

Zdroj: 2024 IEEE Sensors Applications Symposium (SAS). :1-6

Predmety: emergency signals, microcontroller, Speech enhancement, assistive technology, edge devices

Privacy-Preserving Artificial Intelligence on Edge Devices: A Homomorphic Encryption Approach

Autori: Muhammad Jahanzeb Khan Bo Fang Gaetano Cimino a ďalší

Zdroj: 2024 IEEE International Conference on Web Services (ICWS). 126:395-405

Predmety: Privacy preservation, AI services, Edge devices, Homomorphic encryption

Prístupová URL adresa: https://hdl.handle.net/11386/4875571

Use of wireless technologies in IoT projects

Autori: Tetiana A. Vakaliuk Oleksandr V. Andreiev Oleksandr F. Dubyna a ďalší

Zdroj: Journal of Edge Computing, Vol 3, Iss 2 (2024)

Predmety: TK7885-7895, IoT, Computer engineering. Computer hardware, wireless sensor network, Electronic computers. Computer science, 0202 electrical engineering, electronic engineering, information engineering, edge devices, QA75.5-76.95, 02 engineering and technology, monitoring system

Prístupová URL adresa: https://doaj.org/article/c896e4ed380e48ff9e873a3b674bca34

Performance Evaluation of YOLOv5 on Edge Devices for Personal Protective Equipment Detection

Autori: Miglionico G. C. Ducange P. Marcelloni F. a ďalší

Zdroj: 2024 International Joint Conference on Neural Networks (IJCNN). :1-8

Predmety: Edge Devices, Industrial Safety, Object Detection, Personal Protective Equipment