检索结果-内蒙古大学图书馆

From Rule-Based Models to Deep Learning Transformers Architectures for Natural Language Processing and Sign Language Translation systems: Survey, Taxonomy and Performance Evaluation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Shahin, Nada Ismail, Leila Intelligent Distributed Computing and Systems Lab Department of Computer Science and Software Engineering College of Information Technology United Arab Emirates University United Arab Emirates Emirates Center for Mobility Research United Arab Emirates University Abu Dhabi United Arab Emirates

With the growing Deaf and Hard of Hearing population worldwide and the persistent shortage of certified sign language interpreters, there is a pressing need for an efficient, signs-driven, integrated end-to-end translation system, from sign to gloss to text and vice-versa. There has been a wealth of research on machine translations and related reviews. However, there are few works on sign language machine translation considering the particularity of the language being continuous and dynamic. This paper aims to address this void, providing a retrospective analysis of the temporal evolution of sign language machine translation algorithms and a taxonomy of the Transformers architectures, the most used approach in language translation. We also present the requirements of a real-time Quality-of-Service sign language machine translation system underpinned by accurate deep learning algorithms. We propose future research directions for sign language translation *** Codes I.2, I.2.7, I.4, I.4.9 © 2024, CC BY.

关键词： Neural machine translation

DQFormer: Towards Unified LiDAR Panoptic Segmentation with Decoupled Queries

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Yang, Yu Mei, Jianbiao Liu, Liang Du, Siliang Xiao, Yilin Ra, Jongwon Liu, Yong Xu, Xiao Wu, Huifeng The Institute of Cyber-Systems and Control Zhejiang University Hangzhou China Huawei Technologies Co. Ltd. Wuhan China The Institute of Industrial Technology Research Zhejiang University Hangzhou China The Institute of Intelligent and Software Technology Hangzhou Dianzi University Hangzhou China

LiDAR panoptic segmentation, which jointly performs instance and semantic segmentation for things and stuff classes, plays a fundamental role in LiDAR perception tasks. While most existing methods explicitly separate these two segmentation tasks and utilize different branches (i.e., semantic and instance branches), some recent methods have embraced the query-based paradigm to unify LiDAR panoptic segmentation. However, the distinct spatial distribution and inherent characteristics of objects(things) and their surroundings(stuff) in 3D scenes lead to challenges, including the mutual competition of things/stuff and the ambiguity of classification/segmentation. In this paper, we propose decoupling things/stuff queries according to their intrinsic properties for individual decoding and disentangling classification/segmentation to mitigate ambiguity. To this end, we propose a novel framework dubbed DQFormer to implement semantic and instance segmentation in a unified workflow. Specifically, we design a decoupled query generator to propose informative queries with semantics by localizing things/stuff positions and fusing multi-level BEV embeddings. Moreover, a query-oriented mask decoder is introduced to decode corresponding segmentation masks by performing masked cross-attention between queries and mask embeddings. Finally, the decoded masks are combined with the semantics of the queries to produce panoptic results. Extensive experiments on nuScenes and SemanticKITTI datasets demonstrate the superiority of our DQFormer framework. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantic Segmentation

DA-Ada: Learning Domain-Aware Adapter for Domain Adaptive Object Detection

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Li, Haochen Zhang, Rui Yao, Hantao Zhang, Xin Hao, Yifan Song, Xinkai Li, Xiaqing Zhao, Yongwei Li, Ling Chen, Yunji Intelligent Software Research Center Institute of Software CAS Beijing China State Key Lab of Processors Institute of Computing Technology CAS Beijing China State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation CAS Beijing China University of Chinese Academy of Sciences Beijing China

Domain adaptive object detection (DAOD) aims to generalize detectors trained on an annotated source domain to an unlabelled target domain. As the visual-language models (VLMs) can provide essential general knowledge on unseen images, freezing the visual encoder and inserting a domain-agnostic adapter can learn domain-invariant knowledge for DAOD. However, the domain-agnostic adapter is inevitably biased to the source domain. It discards some beneficial knowledge discriminative on the unlabelled domain, ***-specific knowledge of the target domain. To solve the issue, we propose a novel Domain-Aware Adapter (DA-Ada) tailored for the DAOD task. The key point is exploiting domain-specific knowledge between the essential general knowledge and domain-invariant knowledge. DA-Ada consists of the Domain-Invariant Adapter (DIA) for learning domain-invariant knowledge and the Domain-Specific Adapter (DSA) for injecting the domain-specific knowledge from the information discarded by the visual encoder. Comprehensive experiments over multiple DAOD tasks show that DA-Ada can efficiently infer a domain-aware visual encoder for boosting domain adaptive object detection. Our code is available at https://***/Therock90421/DA-Ada © 2024, CC BY-NC-SA.

关键词： Ada (programming language)

Forming Circle Formations of Multi-agent systems Under Actuator Faults 5th

学校读者我要写书评

暂无评论

Forming Circle Formations of Multi-agent Systems Under Actu...

5th Chinese Conference on Swarm Intelligence and Cooperative Control, CCSICC 2021

作者： Liu, Jiajun Ru, Xinfeng Xia, Weiguo Wang, Chen China National Aeronautical Radio Electronic Research Institute and National Key Laboratory of Science and Technology on Avionics System Integration Shanghai China Key Laboratory of Intelligent Control and Optimization for Industrial Equipment of Ministry of Education and School of Control Science and Engineering Dalian University of Technology Dalian China National Engineering Research Center of Software Engineering and the State Key Laboratory of Turbulence and Complex Systems Intelligent Biomimetic Design Lab College of Engineering Peking University Beijing China

ISBN: (纸本)9789811939976

A distributed fault-tolerant formation control law is designed in this paper for multi-agent systems under external disturbances and actuator faults including time-varying loss of effectiveness faults. The initial position of each agent is already on a circle and the control purpose is to form a desired formation on this circle. Based on this local information, a distributed adaptive formation control law is designed by employing the congelation of variables method to deal with time-varying actuator faults. Finally, a simulation example is given to verify the effectiveness of the adaptive fault-tolerant control design method. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Multi agent systems

Dynamic Image Deblurring based on Crosshatch Attention Adversarial Network and Hybrid Loss

学校读者我要写书评

暂无评论

Dynamic Image Deblurring based on Crosshatch Attention Adver...

International Conference on Advancements in Computational Sciences (ICACS)

作者： Ummama Aslam Arslan Shaukat Basim Azam Dept. of Computer and Software Engineering College of Electrical and Mechanical Engineering (CEME) National University of Sciences and Technology (NUST) Islamabad Pakistan Institute of Integrated and Intelligent Systems Griffith University Brisbane Australia

Image deblurring techniques that uses deep learning have shown great potential but due to low generalizability, noise immunity and the correlation among different pixels is not addressed in detail that results in unwanted artifact that appears in the deblurred image. To tackle this problem an end-to-end approach is proposed for the recovery of sharp image from blurred one without the estimation of blur kernel. A special type of attention module known as crosshatch attention is used after Residual Block of Generator model for removing noise and for the collection of correlation of different pixels in an image. Hybrid Loss function is defined which focus on different part of image and improve edges and texture details. The performance of the model for deblurring is measured on GoPro dataset. Our proposed model has slightly higher objective and subjective evaluation i-e PSNR, SSIM value and the visual results.

关键词：

Truncated Cauchy Loss for Outlier-Robust Blind Image Deblurring

学校读者我要写书评

暂无评论

Truncated Cauchy Loss for Outlier-Robust Blind Image Deblurr...

Chinese Control and Decision Conference, CCDC

作者： Xiaopan Li Shiqian Wu Qi Wu Institute of Robotics and Intelligent Systems School of Information Science and Engineering Wuhan University of Science and Technology Wuhan China School of Software & Internet of Things Engineering Jiangxi University of Finance & Economics Nanchang China

ISBN: (数字)9798350387780

ISBN: (纸本)9798350387797

In complex imaging environments, captured images often contain outliers like impulse noise, Cauchy noise, and saturated pixels, posing significant challenges for image deblurring. While existing blind deblurring methods can address outliers to a certain extent, the incorporation of intricate operations tends to amplify the complexity of the deblurring process. In light of these challenges, we present a simple yet robust blind deblurring algorithm for handling images with outliers. To alleviate the effect of outliers on kernel estimation, we introduce a truncated Cauchy loss function, which effectively suppresses outliers by truncating significant errors. Furthermore, we employ a statistics-based approach for outlier detection, eliminating the need for heuristic methods. The proposed approach is shown to be easily implementable and extendable for non-blind deblurring. Experimental results demonstrate that the proposed algorithm outperforms state-of-the-art methods for both blurred images with outliers and those without.

关键词： Noise Imaging Estimation Linear programming Data models Image restoration Complexity theory

An Adaptive Decision-Making Approach for Better Selection of Blockchain Platform for Health Insurance Frauds Detection with Smart Contracts: Development and Performance Evaluation

学校读者我要写书评

暂无评论

Procedia Computer Science 2023年 220卷 470-477页

作者： Rima Kaafarani Leila Ismail Oussama Zahwe ICCS-Lab Computer Science Department American University of Culture and Education 1507 Beirut Lebanon Cloud Computing and Distributed Systems (CLOUDS) Laboratory School of Computing and Information Systems The University of Melbourne Melbourne Australia Intelligent Distributed Computing and Systems (INDUCE) Laboratory Department of Computer Science and Software Engineering College of Information Technology United Arab Emirates University Abu Dhabi United Arab Emirates National Water and Energy Center United Arab Emirates University Abu Dhabi United Arab Emirates

Blockchain technology has piqued the interest of businesses of all types, while consistently improving and adapting to business requirements. Several blockchain platforms have emerged, making it challenging to select a suitable one for a specific type of business. This paper presents a classification of over one hundred blockchain platforms. We develop smart contracts for detecting healthcare insurance frauds using the top two blockchain platforms selected based on our proposed decision-making map approach which selects the top suitable platforms for healthcare insurance frauds detection application. Our classification shows that the largest percentage of platforms can be used for all types of application domains, the second biggest percentage for financial services, and a small number is to develop applications in specific domains. Our decision-making map and performance evaluations reveal that Hyperledger Fabric surpassed Neo in all metrics for detecting healthcare insurance frauds.

关键词： Blockchain Decision-Map Health Insurance Frauds Detection Hyperledger Fabric Neo Smart Contract Smart Healthcare

AmgR: Algebraic Multigrid Accelerated on ReRAM

学校读者我要写书评

暂无评论

AmgR: Algebraic Multigrid Accelerated on ReRAM

Design Automation Conference

作者： Mingjia Fan Xiaotian Tian Yintao He Junxian Li Yiru Duan Xiaozhe Hu Ying Wang Zhou Jin Weifeng Liu Super Scientific Software Laboratory China University of Petroleum Beijing Chinese Academy of Sciences Institute of Computing Technology SKLP Institute of Computing Technology CAS Tufts University USA Research Center for Intelligent Computing Systems

Solving systems of linear equations is a fundamental problem in scientific computing, which has been extensively researched for decades. One of the most well-known solvers is Algebraic Multigrid (AMG), which is widely used in high performance computing due to its good scalability. But currently accelerating AMG relies on the traditional von Neumann architecture of storage and computation separation, which leads to a large data transmission overhead. In this work, we propose a ReRAM-based processing-in-memory (PIM) architecture named AmgR, which overcomes the limitations of the traditional von Neumann architecture for AMG ***, accelerating AMG on ReRAM is non-trivial, because (1) AMG has many computing kernels of various types; (2) there are irregular operations that cannot be directly performed using matrix-vector multiplication suitable for ReRAM, i.e., aggregation operation; (3) ReRAM has poor write endurance, and a lot of data during AMG acceleration needs to be rewritten into ReRAM, resulting in high write cost. To address these issues, firstly, we propose a flexible architecture, which can realize each kernel of AMG and is reused by many kernels to improve resource utilization. Secondly, we propose a dedicated unit to realize the aggregation operation. Finally, we present a new mapping strategy to greatly reduce the number of data handling and writes. The experimental results show that the performance of AmgR is improved by an average of one and two orders of magnitude compared to HYPRE on the CPU and AmgX on the GPU, respectively, while the energy consumption is reduced by an average of two and three orders of magnitude.

关键词：

Multi-robot Task Allocation and Path Planning with Maximum Range Constraints

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Xu, Gang Wu, Yuchen Tao, Sheng Yang, Yifan Liu, Tao Huang, Tao Wu, Huifeng Liu, Yong The Institute of Cyber-Systems and Control Zhejiang University Hangzhou310027 China The Polytechnic Institute Zhejiang University Hangzhou310015 China The Ocean College Zhejiang University Zhoushan316021 China The Institute of Intelligent and Software Technology Hangzhou Dianzi University Hangzhou310018 China

This letter presents a novel multi-robot task allocation and path planning method that considers robots' maximum range constraints in large-sized workspaces, enabling robots to complete the assigned tasks within their range limits. Firstly, we developed a fast path planner to solve global paths efficiently. Subsequently, we propose an innovative auction-based approach that integrates our path planner into the auction phase for reward computation while considering the robots' range limits. This method accounts for extra obstacle-avoiding travel distances rather than ideal straight-line distances, resolving the coupling between task allocation and path planning. Additionally, to avoid redundant computations during iterations, we implemented a lazy auction strategy to speed up the convergence of the task allocation. Finally, we validated the proposed method's effectiveness and application potential through extensive simulation and real-world experiments. The implementation code for our method will be available at https://***/wuuya1/RangeTAP. © 2024, CC BY.

关键词： Multipurpose robots