检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Lian, Shijie Zhang, Ziyi Li, Hua Li, Wenjie Yang, Laurence Tianruo Kwong, Sam Cong, Runmin Hainan University China The Hong Kong University of Science and Technology Guangzhou China Key Laboratory of New Generation Artificial Intelligence Technology & Its Interdisciplinary Applications Southeast University Ministry of Education China Huazhong University of Science and Technology China St. Francis Xavier University Canada LIngnan University School of Data Science Hong Kong Shandong University School of Control Science and Engineering China Key Laboratory of Machine Intelligence and System Control Ministry of Education Jinan China

With the breakthrough of large models, Segment Anything Model (SAM) and its extensions have been attempted to apply in diverse tasks of computer vision. Underwater salient instance segmentation is a foundational and vital step for various underwater vision tasks, which often suffer from low segmentation accuracy due to the complex underwater circumstances and the adaptive ability of models. Moreover, the lack of large-scale datasets with pixel-level salient instance annotations has impeded the development of machine learning techniques in this field. To address these issues, we construct the first large-scale underwater salient instance segmentation dataset (USIS10K), which contains 10,632 underwater images with pixel-level annotations in 7 categories from various underwater scenes. Then, we propose an Underwater Salient Instance Segmentation architecture based on Segment Anything Model (USIS-SAM) specifically for the underwater domain. We devise an Underwater Adaptive Visual Transformer (UA-ViT) encoder to incorporate underwater domain visual prompts into the segmentation network. We further design an out-of-the-box underwater Salient Feature Prompter Generator (SFPG) to automatically generate salient prompters instead of explicitly providing foreground points or boxes as prompts in SAM. Comprehensive experimental results show that our USIS-SAM method can achieve superior performance on USIS10K datasets compared to the state-of-the-art methods. Datasets and codes are released on Github. Copyright © 2024, The Authors. All rights reserved.

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

Adaptive Fault-Tolerant Boundary control of a Flexible Marine Riser-vessel System Under Actuator and Structural Failures

Adaptive Fault-Tolerant Boundary Control of a Flexible Marin...

引用

Chinese control and Decision Conference, CCDC

作者： Mengfei Li Dong Zhao Wenjing Ren Darong Huang School of Artificial Intelligence Anhui University Hefei P. R. China Key Laboratory of System Control and Information Processing Ministry of Education Shanghai P. R. China School of Electrical Engineering and Automation Hefei University of Technology Hefei P. R. China

ISBN: (数字)9798350387780

ISBN: (纸本)9798350387797

This paper develops an adaptive fault-tolerant boundary control scheme for riser-vessel systems to deal with actuator failure and structural failure, simultaneously. The marine riser-vessel system dynamics is modeled by a fourth-order partial-ordinary differential equation. An adaptive fault-tolerant boundary control framework is proposed, which together with the designed adaptive update law guarantees the uniformly ultimately boundedness of the closed-loop system. The simulation results show that the developed fault-tolerant control method can still make the system state converge in a small area of zero in the presence of actuator failure and structural failure.

关键词： Vibrations Fault tolerance Actuators Asymptotic stability System dynamics Simulation Fault tolerant systems

来源：评论

学校读者我要写书评

暂无评论

Efficient Multi-Query Oriented Continuous Subgraph Matching

Efficient Multi-Query Oriented Continuous Subgraph Matching

引用

International Conference on Data Engineering

作者： Ziyi Ma Jianye Yang Xu Zhou Guoqing Xiao Jianhua Wang Liang Yang Kenli Li Xuemin Lin School of Artificial Intelligence Hebei University of Technology China Guangxi Key Laboratory of Machine Vision and Intelligent Control Wuzhou University China Cyberspace Institute of Advanced Technology Guangzhou University China Department of New Networks PengCheng Laboratory China College of Computer Science and Electronic Engineering Hunan University China Shenzhen Research Institute Hunan University China Antai College of Economics and Management Shanghai Jiao Tong University China

ISBN: (数字)9798350317152

ISBN: (纸本)9798350317169

Continuous subgraph matching (CSM) is a critical task for analyzing dynamic graphs and has a wide range of applications, such as merchant fraud detection, cyber-attack hunting, and rumor detection. Although many efficient CSM algorithms have been recently proposed, they are mainly designed to process a single query. However, in some application scenarios, multi-query oriented continuous subgraph matching (MQCSM) may be of more interest. To our knowledge, the two existing solutions to MQCSM are outdated due to unsatisfactory performance. In this paper, we propose MQ-Match, an efficient approach to MQCSM. First, we design a compact yet effective index structure CCG, which maintains the local matching result of vertices in the data graph using a directed graph. The directed edges in CCG can be utilized as an effective pruning rule for the subsequent incremental matching algorithm when expanding a partial match. Then, we develop a computation sharing incremental matching algorithm. In specific, a set of matching trees is constructed based on the depth-first search trees of the query graphs. By utilizing CCG, we conduct subgraph matching for the matching tree to collect the incremental matches for the query graphs, where the common structures of query graphs are matched only once. Extensive experiments show that MQ-Match can achieve 3.1x-7071.4x speedup over the competitors, and consumes much less memory under the majority of the experiment settings.

关键词： Heuristic algorithms Memory management Directed graphs Data engineering Fraud Computational efficiency Indexes

来源：评论

学校读者我要写书评

暂无评论

Temporal Action Localization with Multi-temporal Scales

arXiv

引用

arXiv 2022年

作者： Gao, Zan Cui, Xinglei Zhuo, Tao Cheng, Zhiyong Liu, An-An Wang, Meng Chen, Shenyong Shandong Artificial Intelligence Institute Qilu University of Technology Shandong Academy of Sciences Jinan250014 China The School of Electrical and Information Engineering Tianjin University Tianjin300072 China The school of Computer Science and Information Engineering Hefei University of Technology Hefei230009 China Key Laboratory of Computer Vision and System Ministry of Education Tianjin University of Technology Tianjin300384 China

Temporal action localization plays an important role in video analysis, which aims to localize and classify actions in untrimmed videos. The previous methods often predict actions on a feature space of a single-temporal scale. However, the temporal features of a low-level scale lack enough semantics for action classification while a high-level scale cannot provide rich details of the action boundaries. To address this issue, we propose to predict actions on a feature space of multi-temporal scales. Specifically, we use refined feature pyramids of different scales to pass semantics from high-level scales to low-level scales. Besides, to establish the long temporal scale of the entire video, we use a spatial-temporal transformer encoder to capture the long-range dependencies of video frames. Then the refined features with long-range dependencies are fed into a classifier for the coarse action prediction. Finally, to further improve the prediction accuracy, we propose to use a frame-level self attention module to refine the classification and boundaries of each action instance. Extensive experiments show that the proposed method can outperform state-of-the-art approaches on the THUMOS14 dataset and achieves comparable performance on the ActivityNet1.3 dataset. Compared with A2Net (TIP20, Avg{0.3:0.7}), Sub-Action (CSVT2022, Avg{0.1:0.5}), and AFSD (CVPR21, Avg{0.3:0.7}) on the THUMOS14 dataset, the proposed method can achieve improvements of 12.6%, 17.4% and 2.2%, respectively 1 Copyright © 2022, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Fast Coding Mode Prediction for Intra Prediction in VVC SCC

Fast Coding Mode Prediction for Intra Prediction in VVC SCC

引用

IEEE International Conference on Image Processing

作者： Dayong Wang Junyi Yu Xin Lu Frederic Dufaux Hongwei Guo Hui Guo Ce Zhu Chongqing Key Laboratory on Big Data for Bio Intelligence Chongqing University of Posts and Telecommunications Guangxi Key Laboratory of Machine Vision and Intelligent Control Wuzhou University Chongqing Key Laboratory of Image Cognition Chongqing University of Posts and Telecommunications Faculty of Computing Engineering and Media (CEM) De Montfort University Université Paris-Saclay CNRS CentraleSupélec Laboratoire des signaux et systèmes School of Engineering Honghe University Mengzi Yunnan China University of Electronic Science and Technology of China

ISBN: (数字)9798350349399

ISBN: (纸本)9798350349405

Currently, screen content video applications are increasingly widespread in our daily lives. The latest Screen Content Coding (SCC) standard, known as Versatile Video Coding (VVC) SCC, employs screen content Coding Modes (CMs) selection. While VVC SCC achieves high coding efficiency, its coding complexity poses a significant obstacle to the further widespread adoption of screen content video. Hence, it is crucial to enhance the coding speed of VVC SCC. In this paper, we propose a fast mode and splitting decision for Intra prediction in VVC SCC. Specifically, we initially exploit deep learning techniques to predict content types for all CUs. Subsequently, we examine CM distributions of different content types to predict candidate CMs for CUs. We then introduce early skip and early terminate CM decisions for different content types of CUs to further eliminate unlikely CMs. Finally, we develop Block-based Differential Pulse-Code Modulation (BDPCM) early termination to improve coding speed. Experimental results demonstrate that the proposed algorithm can improve coding speed by $34.95 \%$ on average while maintaining almost the same coding efficiency.

关键词： Video coding Deep learning Image coding Modulation Prediction algorithms Complexity theory Copper

来源：评论

学校读者我要写书评

暂无评论

Digital yard remote control solution for cement manufacturing based on Industrial IOT

Digital yard remote control solution for cement manufacturin...

引用

IEEE International Symposium on Parallel and Distributed Processing with Applications and IEEE International Conference on Ubiquitous Computing and Communications (ISPA/IUCC)

作者： Xiongjie Tao Hui Guo Caixu Xu Hubei Normal University Huangshi China Guangzhou ForceControl-Yuanhai IT Co. Ltd Guangdong China Faculty of Arts and Humanities Macao University of Science and Technology Macao China Guangxi Key Laboratory of Machine Vision and Intelligent Control Wuzhou University Wuzhou China

With the development of artificial intelligence and the deep integration of digitalization and intelligent manufacturing. Aiming at the “raw material storage yard for raw material preparation in cement manufacturing” which still adopts the traditional manual control method for production, the realization of the digital storage yard not only improves the production efficiency, reduces the production cost of the enterprise, but also improves the working environment of the employees. Therefore, this paper uses IIoT technology and PLC control technology to digitally transform the site, and uses three-dimensional visualization technology to realize the simulation of intelligent equipment in the yard, and at the same time improves the security protection of production data. The realization shows that the remote control scheme of the digital yard based on the IIoT proposed in this paper is of great significance for the enterprise to realize the goal of intelligent, digital and safe production.

关键词： Solid modeling Computational modeling Process control Production Raw materials Data models Manufacturing

来源：评论

学校读者我要写书评

暂无评论

DriveRP: RAG and Prompt Engineering Embodied Parallel Driving in Cyber-Physical-Social Spaces

DriveRP: RAG and Prompt Engineering Embodied Parallel Drivin...

引用

Digital Twins and Parallel intelligence (DTPI), IEEE International Conference on

作者： Jun Huang Hao Ma Tengchao Zhang Fei Lin Siji Ma Xiao Wang Fei-Yue Wang Faculty of Innovation Engineering Macau University of Science and Technology Macau China School of Artificial Intelligence Anhui University Hefei China State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China

ISBN: (数字)9798350349252

ISBN: (纸本)9798350349269

In recent years, numerous technological advancements in Artificial Generative intelligences (AGIs) have demonstrated significant potential to transform the intelligence acquisition mechanisms in connected autonomous vehicles (CAVs). Integrating technologies like ChatGPT into CAVs can enhance human-machine interactions. However, the emergence of such new traffic entities may introduce unforeseen hallucinations and complex risks that surpass our current understanding. To address these challenges, Retrieval-Augmented Generation (RAG) and prompt engineering technologies are being explored to enhance the reliability and safety of autonomous driving systems. RAG retrieves relevant contextual information, such as driving experiences and real-time road network status, from external databases to ensure that foundation models have access to accurate and timely data for informed decision-making. Prompt engineering optimizes the performance of large language models in autonomous driving systems by designing and refining prompts that guide the models’ responses, thereby improving their relevance and accuracy in various driving scenarios. Together, these technologies enhance the robustness and trustworthiness of autonomous driving systems. This paper proposes DriveRP, a framework that integrates RAG and prompt engineering within the Descriptive-Predictive-Prescriptive intelligence framework of Parallel Driving theory. DriveRP aims to enhance the safety and interpretability of autonomous vehicle trajectory planning, decision-making, and motion control, ultimately achieving the "6S" goals. Grounded in Digital Twins and Metaverse-embodied parallel driving theory, DriveRP provides the infrastructure and foundational intelligence for parallel driving with Multi-modal Large Lange Models(MLLMs). Additionally, the paper discusses future trends and potential research directions, focusing on the "6S" goals of parallel driving: Smart, Safe, Secure, Sensitive, Sustainable, and Serviceable.

关键词： Accuracy Trajectory planning Decision making Transportation Transforms Safety Digital twins Prompt engineering Motion control Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Exposure-NeuS: Multi-View Head Reconstruction Using Exposure-Distorted Data

Exposure-NeuS: Multi-View Head Reconstruction Using Exposure...

引用

Complex Systems and Intelligent Science (CSIS-IAC), International Annual Conference on

作者： Zhanliang Cheng Bo Peng Ning An School of Data Science University of Science and Technology of China Hefei China School of Mathematical Science University of Science and Technology of China Hefei China Research Institute of Mine Artificial Intelligence China Coal Research Institute Beijing China State Key Laboratory of Intelligent Coal Mining and Strata Control Beijing China

ISBN: (数字)9798331504755

ISBN: (纸本)9798331504762

We propose Exposure-NeuS, a high fidelity human head reconstruction method from multi-view images, even from exposure-distorted images. Previous methods either did not reconstruct with sufficient accuracy, such as failing to reconstruct hair, or resulted in artifacts or holes in the reconstructed results due to exposure in the input data. To solve these challenging problems, we improve the original fine sampling of the model by using piecewise power function sampling to obtain sampling points closer to the real surface. And we develop a prior latent code to implicitly model factors such as environment and exposure. Experiments show that we eliminate or reduce the erroneous facial depressions and hair holes that occur in the baseline method during reconstruction. Released code is provided as follows:https://***/zhanliangOO/***

关键词： Hair Surface reconstruction Head Codes Reconstruction algorithms Rendering (computer graphics) Depression Digital humans Complex systems Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Lightweight vision Transformer for Small Data Sets

Lightweight Vision Transformer for Small Data Sets

引用

Youth Academic Annual Conference of Chinese Association of Automation (YAC)

作者： Wenxing Fang Yanxu Su Institutes of Physical Science and Information Technology Anhui University Hefei China School of Artificial Intelligence Anhui University Hefei P. R. China Key Laboratory of Measurement and Control of Complex System of Engineering Ministry of Education Southeast University Nanjing P. R. China

With the rapid development of Transformers in the field of computer vision, models based on Transformers have become highly competitive architectures in the area. Although variants of Transformer models have achieved increasing accuracy on image classification tasks, the size of the training set and the number of parameters required by the models have increased dramatically. When dealing with small datasets, such models face problems such as overfitting and undergeneralization, leading to poor accuracy on the test set. We propose a new lightweight vision transformer (LVT) to address these issues. We reconstructed the backbone network, which learns the relationship between pixels through local window self-attention and global self-attention computation. We also use the attention pooling approach to fuse the token sequences generated by the backbone network more meticulously. We trained on the CIFAR-10 and CIFAR-100 datasets from scratch and compared them with a modern convolutional neural network. The experimental results show that LVT outperforms the modern convolutional neural network in terms of accuracy and efficiency. On the test set of CIFAR-10, we have obtained an accuracy of 96.83%, which indicates that our model can effectively solve the problems facing the training of small datasets and has a wide range of application prospects.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Scalable block-diagonal locality-constrained projective dictionary learning

arXiv

引用

arXiv 2019年

作者： Zhang, Zhao Jiang, Weiming Zhang, Zheng Li, Sheng Liu, Guangcan Qin, Jie School of Computer Science School of Artificial Intelligence Hefei University of Technology China School of Computer Science and Technology Soochow University China School of Information Technology and Electrical Engineering University of Queensland Australia Department of Computer Science University of Georgia United States School of Information and Control Nanjing University of Information Science and Technology China Computer Vision Laboratory ETH Zurich Switzerland

We propose a novel structured discriminative blockdiagonal dictionary learning method, referred to as scalable Locality-Constrained Projective Dictionary Learning (LC-PDL), for efficient representation and classification. To improve the scalability by saving both training and testing time, our LC-PDL aims at learning a structured discriminative dictionary and a block-diagonal representation without using costly l0/l1-norm. Besides, it avoids extra time-consuming sparse reconstruction process with the well-trained dictionary for new sample as many existing models. More importantly, LC-PDL avoids using the complementary data matrix to learn the sub-dictionary over each class. To enhance the performance, we incorporate a locality constraint of atoms into the DL procedures to keep local information and obtain the codes of samples over each class separately. A block-diagonal discriminative approximation term is also derived to learn a discriminative projection to bridge data with their codes by extracting the special block-diagonal features from data, which can ensure the approximate coefficients to associate with its label information clearly. Then, a robust multiclass classifier is trained over extracted block-diagonal codes for accurate label predictions. Experimental results verify the effectiveness of our algorithm. Copyright © 2019, The Authors. All rights reserved.

关键词： machine learning

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：