检索结果-内蒙古大学图书馆

27th International Conference on Pattern Recognition, ICPR 2024

作者： Ji, Zexin Zou, Beiji Kui, Xiaoyan Vera, Pierre Ruan, Su School of Computer Science and Engineering Central South University Changsha410083 China Hunan Engineering Research Center of Machine Vision and Intelligent Medicine Central South University Changsha410083 China University of Rouen-Normandy LITIS - QuantIF UR 4108 Rouen76000 France Department of Nuclear Medicine Henri Becquerel Cancer Center Rouen France

ISBN: (纸本)9783031781940

In this paper, we propose a self-prior guided Mamba-UNet network (SMamba-UNet) for medical image super-resolution. Existing methods are primarily based on convolutional neural networks (CNNs) or Transformers. CNNs-based methods fail to capture long-range dependencies, while Transformer-based approaches face heavy calculation challenges due to their quadratic computational complexity. Recently, State Space Models (SSMs) especially Mamba have emerged, capable of modeling long-range dependencies with linear computational complexity. Inspired by Mamba, our approach aims to learn the self-prior multi-scale contextual features under Mamba-UNet networks, which may help to super-resolve low-resolution medical images in an efficient way. Specifically, we obtain self-priors by perturbing the brightness inpainting of the input image during network training, which can learn detailed texture and brightness information that is beneficial for super-resolution. Furthermore, we combine Mamba with Unet network to mine global features at different levels. We also design an improved 2D-Selective-Scan (ISS2D) module to divide image features into different directional sequences to learn long-range dependencies in multiple directions, and adaptively fuse sequence information to enhance super-resolved feature representation. Both qualitative and quantitative experimental results demonstrate that our approach outperforms current state-of-the-art methods on two public medical datasets: the IXI and fastMRI. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

EZ-CLIP: EFFICIENT ZERO-SHOT VIDEO ACTION RECOGNITION

arXiv

引用

arXiv 2023年

作者： Ahmad, Shahzad Rawat, Yogesh S. Chanda, Sukalpa Department of Computer Science and Communication Østfold University College Halden Norway Center for Research in Computer Vision University of Central Florida United States

Recent advancements in large-scale pre-training of visual-language models on paired image-text data have demonstrated impressive generalization capabilities for zero-shot tasks. Building on this success, efforts have been made to adapt these image-based visual-language models, such as CLIP, for videos extending their zero-shot capabilities to the video domain. While these adaptations have shown promising results, they come at a significant computational cost and struggle with effectively modeling the crucial temporal aspects inherent to the video domain. In this study, we present EZ-CLIP, a simple and efficient adaptation of CLIP that addresses these challenges. EZ-CLIP leverages temporal visual prompting for seamless temporal adaptation, requiring no fundamental alterations to the core CLIP architecture while preserving its remarkable generalization abilities. Moreover, we introduce a novel learning objective that guides the temporal visual prompts to focus on capturing motion, thereby enhancing its learning capabilities from video data. We conducted extensive experiments on five different benchmark datasets, thoroughly evaluating EZ-CLIP for zero-shot learning and base-to-novel video action recognition, and also demonstrating its potential for few-shot generalization. Impressively, with a mere 5.2 million learnable parameters (as opposed to the 71.1 million in the prior best model), EZ-CLIP can be efficiently trained on a single GPU, outperforming existing approaches in several evaluations. For the latest updates and access to the code, please check the supplementary materials.1 © 2023, CC BY-NC-ND.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

The Extended Optimal Control Problem and Numerical Techniques of Its Solving

The Extended Optimal Control Problem and Numerical Technique...

引用

International Conference on Control, Decision and Information Technologies (CoDIT)

作者： Askhat Diveev Artem Barabash Department of Robotics Control Federal Research Center “Computer Science and Control” Russian Academy of Sciences Moscow Russia Federal Research Center “Computer Science and Control” of the Russian Academy of Sciences Moscow Russia

A new numerical method for solving the optimal control problem in class practically implemented solutions is presented. The method uses an approach of the synthesized control and takes account uncertainties of initial states. Like as synthesized control the method move a control object changing location of stable equilibrium point. As a result it chooses from all possible optimal synthesized controls such, that less sensitive to changes of initial states. As an example, the optimal control problem of quadcopter with complex phase constraints in the form of obstacle areas and narrow bottle neck is considered. To solve this problem firstly the synthesis control problem is solved for obtaining stable equilibrium point in the state space by symbolic regression. After that positions of stable equilibrium points are searched according to source functional from the optimal control problem. As an example, the optimal control problem of quadcopter with complex phase constraints in the form of obstacle areas and narrow bottle neck is considered. To solve this problem firstly the synthesis control problem is solved for obtaining stable equilibrium point in the state space by symbolic regression. After that positions of stable equilibrium points are searched according to source functional from the optimal control problem. Additionally at the search of equilibrium point positions the goal functional is calculated as a sum of fuctional values for all given points of initial states.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Numerical Method for Complete Solution of the Optimal Control Problem

Numerical Method for Complete Solution of the Optimal Contro...

引用

International Conference on Control, Decision and Information Technologies (CoDIT)

作者： Askhat Diveev Department of Robotics Control Federal Research Center “Computer Science and Control” of the Russian Academy of Sciences Moscow Russia

The work is devoted to the numerical complete solution of the optimal control problem. The complete solution means the solution of the optimal control problem together with the solution of the control synthesis problem to stabilize the movement of the control object along the found optimal trajectory. To solve this problem, evolutionary computations and symbolic regression are used. First, the optimal control problem by an evolutionary algorithm in the classical formulation is solved, after that the control synthesis problem by a method of symbolic regression is solved. The statement of the complete optimal control problem is presented. The computational experiment considers the solution of the complete optimal control problem for a quadcopter.

关键词：

来源：评论

学校读者我要写书评

暂无评论

MassSpecGym: A benchmark for the discovery and identification of molecules 38

MassSpecGym: A benchmark for the discovery and identificatio...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Bushuiev, Roman Bushuiev, Anton de Jonge, Niek F. Young, Adamo Kretschmer, Fleming Samusevich, Raman Heirman, Janne Wang, Fei Zhang, Luke Dührkop, Kai Ludwig, Marcus Haupt, Nils A. Kalia, Apurva Brungs, Corinna Schmid, Robin Greiner, Russell Wang, Bo Wishart, David S. Liu, Li-Ping Rousu, Juho Bittremieux, Wout Rost, Hannes Mak, Tytus D. Hassoun, Soha Huber, Florian van der Hooft, Justin J.J. Stravs, Michael A. Böcker, Sebastian Sivic, Josef Pluskal, Tomáš Institute of Organic Chemistry and Biochemistry The Czech Academy of Sciences Czech Republic Czech Institute of Informatics Robotics and Cybernetics Czech Technical University Czech Republic Bioinformatics Group Wageningen University & Research Netherlands Department of Computer Science University of Toronto Canada Bioinformatics Institute for Computer Science Friedrich Schiller University Jena Germany Department of Computer Science University of Antwerp Belgium Department of computing science University of Alberta Canada Alberta Machine Intelligence Institute Canada Department of Molecular Genetics University of Toronto Canada Bright Giant GmbH Department of Computer Science Tufts University United States Department of Biological Sciences University of Alberta Canada Department of Computer Science Aalto University Finland Mass Spectrometry Data Center National Institute of Standards and Technology United States Department of Chemical and Biological Engineering Tufts University United States Centre for Digitalisation and Digitality University of Applied Sciences Düsseldorf Germany Department of Biochemistry University of Johannesburg South Africa Eawag Swiss Federal Institute of Aquatic Science and Technology Switzerland

The discovery and identification of molecules in biological and environmental samples is crucial for advancing biomedical and chemical sciences. Tandem mass spectrometry (MS/MS) is the leading technique for high-throughput elucidation of molecular structures. However, decoding a molecular structure from its mass spectrum is exceptionally challenging, even when performed by human experts. As a result, the vast majority of acquired MS/MS spectra remain uninterpreted, thereby limiting our understanding of the underlying (bio)chemical processes. Despite decades of progress in machine learning applications for predicting molecular structures from MS/MS spectra, the development of new methods is severely hindered by the lack of standard datasets and evaluation protocols. To address this problem, we propose MassSpecGym - the first comprehensive benchmark for the discovery and identification of molecules from MS/MS data. Our benchmark comprises the largest publicly available collection of high-quality labeled MS/MS spectra and defines three MS/MS annotation challenges: de novo molecular structure generation, molecule retrieval, and spectrum simulation. It includes new evaluation metrics and a generalization-demanding data split, therefore standardizing the MS/MS annotation tasks and rendering the problem accessible to the broad machine learning community. MassSpecGym is publicly available at https://***/pluskal-lab/MassSpecGym. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Exploiting Aggregation and Segregation of Representations for Domain Adaptive Human Pose Estimation

arXiv

引用

arXiv 2024年

作者： Peng, Qucheng Zheng, Ce Ding, Zhengming Wang, Pu Chen, Chen Center for Research in Computer Vision University of Central Florida Orlando United States Robotics Institute Carnegie Mellon University Pittsburgh United States Department of Computer Science Tulane University New Orleans United States University of North Carolina at Charlotte Charlotte United States

Human pose estimation (HPE) has received increasing attention recently due to its wide application in motion analysis, virtual reality, healthcare, etc. However, it suffers from the lack of labeled diverse real-world datasets due to the time- and labor-intensive annotation. To cope with the label deficiency issue, one common solution is to train the HPE models with easily available synthetic datasets (source) and apply them to real-world data (target) through domain adaptation (DA). Unfortunately, prevailing domain adaptation techniques within the HPE domain remain predominantly fixated on effecting alignment and aggregation between source and target features, often sidestepping the crucial task of excluding domain-specific representations. To rectify this, we introduce a novel framework that capitalizes on both representation aggregation and segregation for domain adaptive human pose estimation. Within this framework, we address the network architecture aspect by disentangling representations into distinct domain-invariant and domain-specific components, facilitating aggregation of domain-invariant features while simultaneously segregating domain-specific ones. Moreover, we tackle the discrepancy measurement facet by delving into various keypoint relationships and applying separate aggregation or segregation mechanisms to enhance alignment. Extensive experiments on various benchmarks, e.g., Human3.6M, LSP, H3D, and FreiHand, show that our method consistently achieves state-of-the-art performance. The project is available at https: //***/davidpengucf/EPIC. © 2024, CC BY.

关键词： Virtual reality

来源：评论

学校读者我要写书评

暂无评论

Robin: A Novel Method to Produce Robust Interpreters for Deep Learning-Based Code Classifiers 23

Robin: A Novel Method to Produce Robust Interpreters for Dee...

引用

Proceedings of the 38th IEEE/ACM International Conference on Automated Software Engineering

作者： Zhen Li Ruqian Zhang Deqing Zou Ning Wang Yating Li Shouhuai Xu Chen Chen Hai Jin School of Cyber Science and Engineering Huazhong University of Science and Technology Wuhan China Department of Computer Science University of Colorado Colorado Springs USA Center for Research in Computer Vision University of Central Florida USA School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9798350329964

Deep learning has been widely used in source code classification tasks, such as code classification according to their functionalities, code authorship attribution, and vulnerability detection. Unfortunately, the black-box nature of deep learning makes it hard to interpret and understand why a classifier (i.e., classification model) makes a particular prediction on a given example. This lack of interpretability (or explainability) might have hindered their adoption by practitioners because it is not clear when they should or should not trust a classifier's prediction. The lack of interpretability has motivated a number of studies in recent years. However, existing methods are neither robust nor able to cope with out-of-distribution examples. In this paper, we propose a novel method to produce Robust interpreters for a given deep learning-based code classifier; the method is dubbed Robin. The key idea behind Robin is a novel hybrid structure combining an interpreter and two approximators, while leveraging the ideas of adversarial training and data augmentation. Experimental results show that on average the interpreter produced by Robin achieves a 6.11% higher fidelity (evaluated on the classifier), 67.22% higher fidelity (evaluated on the approximator), and 15.87x higher robustness than that of the three existing interpreters we evaluated. Moreover, the interpreter is 47.31% less affected by out-of-distribution examples than that of LEMNA.

关键词： explainable AI

来源：评论

学校读者我要写书评

暂无评论

Lumbar spine localisation method based on feature fusion

引用

CAAI Transactions on Intelligence Technology 2023年第3期8卷 931-945页

作者： Yonghong Zhang Ning Hu Zhuofu Li Xuquan Ji Shanshan Liu Youyang Sha Xiongkang Song Jian Zhang Lei Hu Weishi Li Robotics Institute School of Mechanical Engineering and AutomationBeihang UniversityBeijingChina Beijing Zhuzheng Robot Co. LTDBeijingChina Department of Mechanical Aerospace and Biomedical EngineeringUniversity of TennesseeKnoxvilleTennesseeUSA Department of Orthopaedics Peking University Third HospitalBeijingChina Engineering Research Center of Bone and Joint Precision Medicine Ministry of EducationBeijingChina Beijing Key Laboratory of Spinal Disease Research BeijingChina School of Biological Science and Medical Engineering Beihang UniversityBeijingChina Department of Computer Science University of WarwickCoventryUK

To eliminate unnecessary background information,such as soft tissues in original CT images and the adverse impact of the similarity of adjacent spines on lumbar image segmentation and surgical path planning,a two‐stage approach for localising lumbar segments is ***,based on the multi‐scale feature fusion technology,a non‐linear regression method is used to achieve accurate localisation of the overall spatial region of the lumbar spine,effectively eliminating useless background information,such as soft *** the second stage,we directly realised the precise positioning of each segment in the lumbar spine space region based on the non‐linear regression method,thus effectively eliminating the interference caused by the adjacent *** 3D Intersection over Union(3D_IOU)is used as the main evaluation indicator for the positioning *** an open dataset,3D_IOU values of 0.8339�0.0990 and 0.8559�0.0332 in the first and second stages,respectively is *** addition,the average time required for the proposed method in the two stages is 0.3274 and 0.2105 s ***,the proposed method performs very well in terms of both pre-cision and speed and can effectively improve the accuracy of lumbar image segmentation and the effect of surgical path planning.

关键词： CT image lumbar spatial orientation multi‐scale information fusion

来源：评论

学校读者我要写书评

暂无评论

Comprehensive Overview of Reward Engineering and Shaping in Advancing Reinforcement Learning Applications

arXiv

引用

arXiv 2024年

作者： Ibrahim, Sinan Mostafa, Mostafa Jnadi, Ali Salloum, Hadi Osinenko, Pavel Skolkovo Institute of Science and Technology Moscow121205 Russia Institute of Robotics and Computer Vision Innopolis University Innopolis420500 Russia Research Center for Artificial Intelligence Innopolis University Innopolis420500 Russia

Reinforcement Learning (RL) seeks to develop systems capable of autonomous decision-making by learning through interaction with their environment. Central to this process are reward engineering and reward shaping, which are essential for enhancing the efficiency and effectiveness of RL algorithms. These techniques guide agents toward desired behaviors, improve learning stability, and accelerate convergence by addressing challenges such as sparse and delayed rewards. However, the complexity of real-world environments and the computational demands of RL algorithms remain significant obstacles to broader adoption. Recent advancements in deep learning have enabled RL to handle high-dimensional state and action spaces, facilitating applications in robotics, autonomous driving, and complex decision-making tasks. In response to these developments, this paper provides one of the first comprehensive reviews of reward design in RL, with a focus on the methodologies and techniques underpinning reward engineering and shaping. By introducing a detailed taxonomy, critically analyzing current approaches, and highlighting their limitations, this work fills an important gap in the literature, offering insights into how reward structures can be optimized to meet the growing demands of modern AI systems. © 2024, CC BY.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Segmentation and Rigid Registration of Liver Dynamic Computed Tomography Images for Diagnostic Assessment of Fatty Liver Disease

引用

Journal of Computing science and Engineering 2023年第3期17卷 117-126页

作者： Koo, Kyoyeong Lee, Jeongjin Hwang, Jiwon Park, Taeyong Jeong, Heeryeol Khang, Seungwoo Lee, Jongmyoung Kwon, Hyuk Na, Seungwon Lee, Sunyoung Kim, Kyoung Won Kim, Kyung Won School of Computer Science and Engineering Soongsil University Seoul Korea Republic of Hanwha Vision Co. Ltd. Seongnam Korea Republic of Department of Biomedical Informatics Hallym University Medical Center Anyang Korea Republic of SKIA Inc. Seoul Korea Republic of Department of Radiology and Research Institute of Radiological Science Severance Hospital Yonsei University College of Medicine Seoul Korea Republic of Department of Radiology and Research Institute of Radiology Asan Medical Center University of Ulsan College of Medicine Seoul Korea Republic of iAID Inc. Seoul Korea Republic of

This study presents a method for diagnosing fatty liver disease by using time-difference liver computed tomography (CT) images of the same patient to perform segmentation and rigid registration on liver regions, excluding the vascular regions. The proposed method comprises three main steps. First, the liver region is segmented in the precontrast phase, and the liver and liver vessel regions are segmented in the portal phase. Second, rigid registration is performed between the liver regions to align the liver positions affected by the patient's posture or breathing. Finally, fatty liver diagnosis is performed with the average Hounsfield unit (HU) value calculated using only the area removed from the vessel area segmented in the portal phase after registration in the precontrast liver area. The mean distance error between the points corresponding to the liver boundary was 3.136 mm and the mean error between the anatomic landmarks was 4.166 mm. A fatty liver diagnosis was confirmed in a total of 18 cases, and the results were identical to the histology results. This technique may be valuable in clinically diagnosing fatty liver using liver CT imaging, which is widely available and more commonly used than abdominal magnetic resonance. © 2023. The Korean Institute of Information Scientists and Engineers

关键词： Diagnosis

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：