检索结果-内蒙古大学图书馆

arXiv 2020年

作者： Jin, Beibei Hu, Yu Tang, Qiankun Niu, Jingyu Shi, Zhiping Han, Yinhe Li, Xiaowei Research Center for Intelligent Computing Systems State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences University of Chinese Academy of Sciences Capital Normal University

Video prediction is a pixel-wise dense prediction task to infer future frames based on past frames. Missing appearance details and motion blur are still two major problems for current predictive models, which lead to image distortion and temporal inconsistency. In this paper, we point out the necessity of exploring multi-frequency analysis to deal with the two problems. Inspired by the frequency band decomposition characteristic of Human Vision System (HVS), we propose a video prediction network based on multi-level wavelet analysis to deal with spatial and temporal information in a unified manner. Specifically, multi-level spatial discrete wavelet transform decomposes each video frame into anisotropic sub-bands with multiple frequencies, helping to enrich structural information and reserve fine details. On the other hand, multi-level temporal discrete wavelet transform which operates on time axis decomposes the frame sequence into sub-band groups of different frequencies to accurately capture multi-frequency motions under fixed frame rate. Extensive experiments on diverse datasets demonstrate that our model shows significant improvements on fidelity and temporal consistency over the state-of-the-art works. Copyright © 2020, The Authors. All rights reserved.

关键词： Discrete wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

architecture and key Technologies of Parallel Dispatching System for Railway Technical Operation Stations

Architecture and Key Technologies of Parallel Dispatching Sy...

引用

Complex systems and intelligent Science (CSIS-IAC), International Annual Conference on

作者： Gang Xiong Donghu Yang Wei Xu Runmei Li Shichao Chen Bing Song Xisong Dong Fenghua Zhu State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences Beijing China Beijing Engineering Research Center of Intelligent Systems and Technology Institute of Automation Chinese Academy of Sciences Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China Signal and Communication Research Institute China Academy of Railway Science Corporation Limited Beijing China The School of Electronic and Information Engineering Beijing Jiaotong University Beijing China Beijing Huairou Academy of Parallel Sensing Beijing China Guangdong Engineering Research Center of 3D Printing and Intelligent Manufacturing Cloud Computing Center Chinese Academy of Sciences Dongguan China

A railroad technical operation station is an extensive and complex operation system and a variety of random disturbing factors further increase the complexity of its command and dispatching. To enhance the railroad technical operation stations' intelligence, and integration level, and enhance its dispatching management ability of complex operation plans, based on the ACP approach, this paper designs the parallel dispatching system (PDS) architecture, and describes its key technologies including the artificial dispatching system (ADS), computing experiments, parallel execution, etc. On the basis of the artificial dispatching system, computational experiments can optimize various types of technical operations and the overall operational process, to achieve parallel execution between the actual dispatching system and ADS. PDS can provide more efficient and intelligent command and dispatching solutions for railroad technical work stations, and promote the development of the railroad transportation industry in the direction of higher quality and efficiency.

关键词：

来源：评论

学校读者我要写书评

暂无评论

GSLB: The Graph Structure Learning Benchmark

arXiv

引用

arXiv 2023年

作者： Li, Zhixun Wang, Liang Sun, Xin Luo, Yifan Zhu, Yanqiao Chen, Dingshuo Luo, Yingtao Zhou, Xiangxin Liu, Qiang Wu, Shu Yu, Jeffrey Xu Department of Systems Engineering and Engineering Management The Chinese University of Hong Kong Hong Kong Center for Research on Intelligent Perception Computing State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences China School of Artificial Intelligence University of Chinese Academy of Sciences China Department of Automation University of Science and Technology of China China School of Cyberspace Security Beijing University of Posts and Telecommunications China Department of Computer Science University of California Los Angeles United States Heinz College of Information Systems and Public Policy Machine Learning Department School of Computer Science Carnegie Mellon University United States

Graph Structure Learning (GSL) has recently garnered considerable attention due to its ability to optimize both the parameters of Graph Neural Networks (GNNs) and the computation graph structure simultaneously. Despite the proliferation of GSL methods developed in recent years, there is no standard experimental setting or fair comparison for performance evaluation, which creates a great obstacle to understanding the progress in this field. To fill this gap, we systematically analyze the performance of GSL in different scenarios and develop a comprehensive Graph Structure Learning Benchmark (GSLB) curated from 20 diverse graph datasets and 16 distinct GSL algorithms. Specifically, GSLB systematically investigates the characteristics of GSL in terms of three dimensions: effectiveness, robustness, and complexity. We comprehensively evaluate state-of-the-art GSL algorithms in node- and graph-level tasks, and analyze their performance in robust learning and model complexity. Further, to facilitate reproducible research, we have developed an easy-to-use library for training, evaluating, and visualizing different GSL methods. Empirical results of our extensive experiments demonstrate the ability of GSL and reveal its potential benefits on various downstream tasks, offering insights and opportunities for future research. The code of GSLB is available at: https://***/GSL-Benchmark/GSLB. © 2023, CC BY.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

S$^{\text{4}}$TP: Social-Suitable and Safety-Sensitive Trajectory Planning for Autonomous Vehicles

IEEE Transactions on Intelligent Vehicles

引用

IEEE Transactions on intelligent Vehicles 2023年第2期9卷 3220-3231页

作者： Xiao Wang Ke Tang Xingyuan Dai Jintao Xu Quancheng Du Rui Ai Yuxiao Wang Weihao Gu Engineering Research Center of Autonomous Unmanned System Technology Ministry of Education Anhui University Hefei China Haomo Technology Company Ltd. Beijing China State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China Qingdao Academy of Intelligent Industries Qingdao China School of Computer and Communication Engineering University of Science and Technology Beijing Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China

In public roads, autonomous vehicles (AVs) face the challenge of frequent interactions with human-driven vehicles (HDVs), which render uncertain driving behavior due to varying social characteristics among humans. To effectively assess the risks prevailing in the vicinity of AVs in social interactive traffic scenarios and achieve safe autonomous driving, this article proposes a social-suitable and safety-sensitive trajectory planning (S $^{\text{4}}$ TP) framework. Specifically, S $^{\text{4}}$ TP integrates the Social-Aware Trajectory Prediction (SATP) and Social-Aware Driving Risk Field (SADRF) modules. SATP utilizes Transformers to effectively encode the driving scene and incorporates an AV's planned trajectory during the prediction decoding process. SADRF assesses the expected surrounding risk degrees during AVs-HDVs interactions, each with different social characteristics, visualized as two-dimensional heat maps centered on the AV. SADRF models the driving intentions of the surrounding HDVs and predicts trajectories based on the representation of vehicular interactions. S $^{\text{4}}$ TP employs an optimization-based approach for motion planning, utilizing the predicted HDVs' trajectories as input. With the integration of SADRF, S $^{\text{4}}$ TP executes real-time online optimization of the planned trajectory of AV within low-risk regions, thus improving the safety and the interpretability of the planned trajectory. We have conducted comprehensive tests of the proposed method using the SMARTS simulator. Experimental results in complex social scenarios, such as unprotected left-turn intersections, merging, cruising, and overtaking, validate the superiority of our proposed S $^{\text{4}}$ TP in terms of safety and rationality. S $^{\text{4}}$ TP achieves a pass rate of 100% across all scenarios, surpassing the current state-of-the-art methods Fanta of 98.25% and Predictive-Decision of 94.75%.

关键词： Trajectory Safety Vehicles Trajectory planning Autonomous vehicles Risk management Behavioral sciences

来源：评论

学校读者我要写书评

暂无评论

Lightdet: A Lightweight and Accurate Object Detection Network

Lightdet: A Lightweight and Accurate Object Detection Networ...

引用

IEEE International Conference on Acoustics, Speech and Signal Processing

作者： Qiankun Tang Jie Li Zhiping Shi Yu Hu Research Center for Intelligent Computing Systems State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences Beijing Advanced Innovation Center for Imaging Theory and Technology Capital Normal University

ISBN: (数字)9781509066315

ISBN: (纸本)9781509066322

The extensive computational burden limits the usage of accurate but complex object detectors in resource-bounded scenarios. In this paper, we present a lightweight object detector, named LightDet, to address this dilemma. We design a lightweight backbone that is able to capture rich low-level features by the proposed Detail-Preserving Module. To effectively aggregate bottom and top-down features, we introduce an efficient Feature-Preserving and Refinement Module. A lightweight prediction head is employed to further reduce the entire network complexity. Experimental results show that our LightDet achieves 75.5% mAP on PASCAL VOC 2007 at the speed of 250 FPS and 24.0% mAP on MS COCO dataset.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Graph Confidence Intercalibration Network for Intracranial Aneurysm Lesion Instance Segmentation in Dsa

SSRN

引用

SSRN 2024年

作者： Ye, Haili Mo, Yancheng Tang, Chen Liao, Mingqian Zhang, Xiaoqing Dai, Limeng Li, Baihua Liu, Jiang Research Institute of Trustworthy Autonomous Systems Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen518055 China Department of Computer Science Loughborough University LoughboroughLE11 3TU United Kingdom School of Ophthalmology and Optometry Eye Hospital Wenzhou Medical University Wenzhou325027 China Center for High Performance Computing and Shenzhen Key Laboratory of Intelligent Bioinformatics Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen518055 China Shenzhen People’s Hospital Neurosurgery Department Shenzhen518020 China China

Intracranial aneurysm (IA) lesion segmentation is significant for its treatment and prognosis. Although exiting deep network-based instance methods have good IA lesion segmentation results based on digital subtraction angiography (DSA) images, they still face great challenges with instance confidence bias and imprecise boundary segmentation, which may negatively affect IA diagnosis. To tackle these problems, this paper proposes a novel graph confidence intercalibration network (GCINet) to automatically segment IA lesions from DSA images. To be specific, we design a graph confidence intercalibration (GCI) module to mitigate instance confidence bias by dynamically adjusting their confidence distributions. At the same time, we propose an edge space perception (ESP) module to correct ambiguous segmentation boundaries. Extensive experiments on a clinical IA-DSA and a publicly available LiTS dataset demonstrate that our GCINet outperforms state-of-the-art methods. Additionally, visual analysis and ablation studies are provided to verify the effectiveness of each module in GCINet. © 2024, The Authors. All rights reserved.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Decentralized Optimization on Compact Submanifolds by Quantized Riemannian Gradient Tracking

引用

IEEE Transactions on Signal Processing 2025年 73卷 1851-1861页

作者： Chen, Jun Liu, Lina Zhu, Tianyi Liu, Yong Dai, Guang Jiang, Yunliang Tsang, Ivor W. Zhejiang Normal University National Special Education Resource Center for Children with Autism Hangzhou311231 China Zhejiang Normal University School of Computer Science and Technology Jinhua321004 China China Mobile Research Institute Beijing100032 China Zhejiang University Institute of Cyber-Systems and Control Hangzhou310027 China State Grid Corporation of China SGIT AI Lab China Zhejiang Normal University Zhejiang Key Laboratory of Intelligent Education Technology and Application Jinhua321004 China Huzhou University School of Information Engineering Huzhou313000 China Centre for Frontier Artificial Intelligence Research A*STAR Singapore

This paper considers the problem of decentralized optimization on compact submanifolds, where a finite sum of smooth (possibly non-convex) local functions is minimized by n agents forming an undirected and connected graph. However, the efficiency of distributed optimization is often hindered by communication bottlenecks. To mitigate this, we propose the Quantized Riemannian Gradient Tracking (Q-RGT) algorithm, where agents update their local variables using quantized gradients. The introduction of quantization noise allows our algorithm to bypass the constraints of the accurate Riemannian projection operator (such as retraction), further improving iterative efficiency. To the best of our knowledge, this is the first algorithm to achieve an O(1/K) convergence rate in the presence of quantization, matching the convergence rate of methods without quantization. Additionally, we explicitly derive lower bounds on decentralized consensus associated with a function of quantization levels. Numerical experiments demonstrate that Q-RGT performs comparably to non-quantized methods while reducing communication bottlenecks and computational overhead. © 1991-2012 IEEE.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

Data-Driven Modeling and Random Search Optimization Method for the Polysilicon Cvd Reactor

SSRN

引用

SSRN 2022年

作者： Xi, Bangwen Xiong, Gang Shen, Zhen Tamir, Tariku Sinshaw He, Chang Song, Yonggang Liu, Xiong State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing100049 China Beijing Engineering Research Center of Intelligent Systems and Technology Institute of Automation Chinese Academy of Sciences Beijing100190 China Guangdong Engineering Research Center of 3D Printing and Intelligent Manufacturing Cloud Computing Center Chinese Academy of Sciences Dongguan523808 China Intelligent Manufacturing Center Qingdao Academy of Intelligent Industries Qingdao266109 China Xinte Energy Co. Ltd Urumqi830000 China

Polysilicon is mainly obtained by chemical vapor deposition(CVD) reaction, in which hydrogen gas and trichlorosilane(TCS) are fed into the CVD reactor to produce polysilicon. However, the feeding parameters are mainly adjusted in a step-by-step manner according to the actual deposition results, which is inefficient to optimize and prone to resource wastage. In this paper, based on the existing mechanistic model of the CVD reactor, we combine the history data with the specific parameters to construct a simulator. A random search algorithm is used to determine the optimal feeding parameters. With the simulator, the polysilicon productivity can be increased by 704 kg, and the power consumption reduced by 3.8 kWh/kg. © 2022, The Authors. All rights reserved.

关键词： Simulators

来源：评论

学校读者我要写书评

暂无评论

research on robot path planning based on multi-strategy genetic ant colony optimization algorithm

引用

Information Sciences 2025年 718卷

作者： Zheng Li Mingyuan Du Jinlei Qin Xuechao Chen Department of Computer North China Electric Power University 071003 Baoding China Hebei Key Laboratory of Knowledge Computing for Energy & Power 071003 Baoding China Engineering Research Center of Intelligent Computing for Complex Energy Systems Ministry of Education 071003 Baoding China

The ant colony optimization (ACO) algorithm is commonly used for path planning but faces issues like slow convergence, a tendency to get trapped in local optima, and instability in performance. This paper presents the multi-strategy genetic ant colony optimization (MSGACO) algorithm, which enhances ACO through several innovations: adaptive pheromone distribution initialization, a three-level node selection mechanism, a direction heuristic function, a selective crossover mechanism, and an elitist selection strategy. The adaptive pheromone distribution initialization, combined with the three-level node selection mechanism and direction heuristic function, increases the selection probability for nodes nearer to the destination. Additionally, the selective crossover mechanism and elitist selection strategy optimize the searched paths, resulting in more high-quality solutions and allocating pheromones to them. The main parameters of these strategies are optimized, and their effectiveness is verified. To showcase the benefits of MSGACO, experiments are performed utilizing five exemplary environmental models, with comparisons made to existing methodologies. The results showed a reduction of approximately 60.7% in path length and 52% in convergence iterations compared to the traditional ant colony optimization algorithm.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Consistency and Diversity induced Human Motion Segmentation

arXiv

引用

arXiv 2022年

作者： Zhou, Tao Fu, Huazhu Gong, Chen Shao, Ling Porikli, Fatih Ling, Haibin Shen, Jianbing The PCA Lab The Key Laboratory of Intelligent Perception and Systems for High-Dimensional Information Ministry of Education School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing210094 China Institute of High Performance Computing Agency for Science Technology and Research Singapore Riyadh Saudi Arabia The Research School of Engineering The Australian National University Australia The Department of Computer Science Stony Brook University Strony Brook NY United States The State Key Laboratory of Internet of Things for Smart City Department of Computer and Information Science University United States

Subspace clustering is a classical technique that has been widely used for human motion segmentation and other related tasks. However, existing segmentation methods often cluster data without guidance from prior knowledge, resulting in unsatisfactory segmentation results. To this end, we propose a novel Consistency and Diversity induced human Motion Segmentation (CDMS) algorithm. Specifically, our model factorizes the source and target data into distinct multi-layer feature spaces, in which transfer subspace learning is conducted on different layers to capture multi-level information. A multi-mutual consistency learning strategy is carried out to reduce the domain gap between the source and target data. In this way, the domain-specific knowledge and domain-invariant properties can be explored simultaneously. Besides, a novel constraint based on the Hilbert Schmidt Independence Criterion (HSIC) is introduced to ensure the diversity of multi-level subspace representations, which enables the complementarity of multi-level representations to be explored to boost the transfer learning performance. Moreover, to preserve the temporal correlations, an enhanced graph regularizer is imposed on the learned representation coefficients and the multi-level representations of the source data. The proposed model can be efficiently solved using the Alternating Direction Method of Multipliers (ADMM) algorithm. Extensive experimental results on public human motion datasets demonstrate the effectiveness of our method against several state-of-the-art approaches. Copyright © 2022, The Authors. All rights reserved.

关键词： Motion analysis

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：