检索结果-内蒙古大学图书馆

26th International Conference on Information Fusion, FUSION 2023

作者： Zhang, Tianhao Wang, Shenglin Bouaynaya, Nidhal Calinescu, Radu Mihaylova, Lyudmila The University of Sheffield Automatic Control and Systems Engineering United Kingdom Rowan University Department of Electrical and Computer Engineering United States University of York Department of Computer Science United Kingdom

ISBN: (纸本)9798890344854

The superior performance of object detectors is often established under the condition that the test samples are in the same distribution as the training data. However, in many practical applications, out-of-distribution (OOD) instances are inevitable and usually lead to uncertainty in the results. In this paper, we propose a novel, intuitive, and scalable probabilistic object detection method for OOD detection. Unlike other uncertainty-modeling methods that either require huge computational costs to infer the weight distributions or rely on model training through synthetic outlier data, our method is able to distinguish between in-distribution (ID) data and OOD data via weight parameter sampling from proposed Gaussian distributions based on pre-trained networks. We demonstrate that our Bayesian object detector can achieve satisfactory OOD identification performance by reducing the FPR95 score by up to 8.19% and increasing the AUROC score by up to 13.94% when trained on BDD100k and VOC datasets as the ID datasets and evaluated on COCO2017 dataset as the OOD dataset. © 2023 International Society of Information Fusion.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

MTP-GO: Graph-Based Probabilistic Multi-Agent Trajectory Prediction with Neural ODEs

IEEE Transactions on Intelligent Vehicles

引用

IEEE Transactions on Intelligent Vehicles 2023年第9期8卷 4223-4236页

作者： Westny, Theodor Oskarsson, Joel Olofsson, Bjorn Frisk, Erik Linköping University Department of Computer and Information Science Linköping581 83 Sweden Lund University Department of Automatic Control Lund221 00 Sweden Linköping University Department of Electrical Engineering Linköping581 83 Sweden

Enabling resilient autonomous motion planning requires robust predictions of surrounding road users' future behavior. In response to this need and the associated challenges, we introduce our model titled MTP-GO. The model encodes the scene using temporal graph neural networks to produce the inputs to an underlying motion model. The motion model is implemented using neural ordinary differential equations where the state-transition functions are learned with the rest of the model. Multimodal probabilistic predictions are obtained by combining the concept of mixture density networks and Kalman filtering. The results illustrate the predictive capabilities of the proposed model across various data sets, outperforming several state-of-the-art methods on a number of metrics. © 2016 IEEE.

关键词： Trajectories

来源：评论

学校读者我要写书评

暂无评论

Image Enhancement for Facial Landmark Detection 21

Image Enhancement for Facial Landmark Detection

引用

21st RoEduNet Conference on Networking in Education and Research, RoEduNet 2022

作者： Ciuntu, Victor Chiornita, Traian Boiangiu, Costin-Anton Tarba, Nicolae Voncila, Mihai-Lucian Rosner, Daniel Politehnica University Faculty of Automatic Control and Computers Computer Science and Engineering Department Bucharest Romania

ISBN: (数字)9781665455282

ISBN: (纸本)9781665455282

Facial landmark detection (FLD) is a field of study in computer vision that specializes in detecting and tracking key points from human faces. There are many applications, such as detecting a human's head pose (position and rotation), tracking whether drivers are paying attention or not, applying augmented reality, etc. This paper proposes to evaluate and decide what are the optimal techniques of image enhancement that can be used to improve different algorithms of FLD. After proposing different pipelines of contrast enhancement and grayscale conversion to enhance the image before applying the FLD algorithm, the results show that the performance will not increase drastically if an image enhancement pipeline is used, but for the reason that this step will not consume a lot of time in the process of detection, it is a step worth considering. © 2022 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Complete OCR Solution for Image Analysis of World War 2 Documents 21

Complete OCR Solution for Image Analysis of World War 2 Docu...

引用

21st RoEduNet Conference on Networking in Education and Research, RoEduNet 2022

作者： Gradinaru, Mihai Negru, Andrei Boiangiu, Costin-Anton Tarba, Nicolae Voncila, Mihai-Lucian Deaconescu, Razvan Adrian Politehnica University Faculty of Automatic Control and Computers Computer Science and Engineering Department Bucharest Romania

ISBN: (数字)9781665455282

ISBN: (纸本)9781665455282

The field of Optical Character Recognition (OCR) consists of techniques that are mainly focused on document image analysis. Aside from generating significant speedups of everyday procedures, OCR has a considerable role in the preservation of historical sources of information. Most of the World War 2 (WW2) documents are of great importance, especially with applications in virtual archives, museums, and research. The situation asks for an efficient, yet not aggressive, transcribing method using OCR tools. This paper describes an approach in the context of the given problem. The focus is oriented towards extracting the information from documents affected by their age, but with simpler structures, mainly split into paragraphs, such as letters and military reports. The approach is based on combining the results of multiple OCR engines, with the final objective of achieving better performance compared to the individual performance of each engine. © 2022 IEEE.

关键词： Engines

来源：评论

学校读者我要写书评

暂无评论

Weighting Attributes Based on the Greedy Algorithm Properties 28th

Weighting Attributes Based on the Greedy Algorithm Propertie...

引用

28th International Conference on Knowledge Based and Intelligent information and Engineering Systems, KES 2024

作者： Zielosko, Beata Stanczyk, Urszula Jablonski, Kamil University of Silesia in Katowice Institute of Computer Science Bedzinska 39 Sosnowiec41-200 Poland Department of Computer Graphics Vision and Digital Systems Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Akademicka 2A Gliwice44-100 Poland

Estimation of importance for considered features is an important issue for any knowledge exploration process and it can be executed by a variety of approaches. In the research reported in this study, the primary aim was the development of a methodology for creating attribute rankings. Based on the properties of the greedy algorithm for inducing decision rules, a new application of this algorithm has been proposed. Instead of constructing a single ordering of features, attributes were weighted multiple times. The input datasets were discretised with several algorithms representing supervised and unsupervised discretisation approaches. Each resulting discrete data variant was exploited to construct a ranking of attributes. The effectiveness of the obtained rankings was confirmed through a rule filtering process governed by weighted attributes. The methodology was applied to the stylometric task of authorship attribution. The experimental outcomes demonstrate the value of the proposed research method, as it generally led to improved predictions while taking into account a noticeably decreased sets of attributes and decision rules. © 2024 The Authors.

关键词： Consensus algorithm

来源：评论

学校读者我要写书评

暂无评论

How transformations of representation for input data can affect the properties of induced decision reducts and rules 27

How transformations of representation for input data can aff...

引用

27th International Conference on Knowledge Based and Intelligent Information and Engineering Sytems, KES 2023

作者： Stanczyk, Urszula Department of Graphics Computer Vision and Digital Systems Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Akademicka 2A Gliwice44-100 Poland

Decision reducts and rules belong to forms used for the representation of knowledge learnt from input data while using a rough set approach in the exploration stage. As with any patterns that capture properties of data, the size is considered as the most important indicator of their quality, as a more general and concise description allows for improved understanding and simplified interpretation. The size of a reduct is given by its cardinality, and for a rule it is a number of included conditions, both referred to as lengths. These lengths undergo transformations along with data transformations, so such processing as discretisation leads to changed characteristics of existing patterns and through them also reducts and rules. The paper presents research with contributions in the observations focused on the influence of supervised and unsupervised discretisation on properties of attributes, induced decision reducts, and rules, offering a deeper insight into relations among them. © 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://***/licenses/by-nc-nd/4.0)

关键词： Data exploration decision reduct decision rule discretisation knowledge representation rough set theory transformation

来源：评论

学校读者我要写书评

暂无评论

Should supervised discretisation always be trusted unreservedly? On combining characteristics of supervised and unsupervised discretisation algorithms in two-step processing 27

Should supervised discretisation always be trusted unreserve...

引用

27th International Conference on Knowledge Based and Intelligent Information and Engineering Sytems, KES 2023

作者： Stanczyk, Urszula Baron, Grzegorz Department of Graphics Computer Vision and Digital Systems Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Akademicka 2A Gliwice44-100 Poland

The paper presents a description of the research methodology dedicated to a two-step discretisation process applied to the input numeric data, with combining the characteristics of selected supervised and unsupervised algorithms, which leads to extended processing of some attributes in train and test sets. The methodology was illustrated with the investigations carried out in the domain of stylometric analysis of texts, for two datasets prepared for the task of binary authorship attribution. The several variants of transformed input data obtained were subjected to exploration using two selected machine learning methods capable of inducing knowledge from both continuous and categorical forms, namely the PART and J48 classifiers. The results from the experiments indicate that, as can be expected, supervised transformations of data work well enough, however, they do not always return the best outcome. The two-step processing of some attributes shows sufficient promise to warrant a closer study, as opposed to always unconditionally relying only on supervised algorithms as outperforming all other approaches. © 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://***/licenses/by-nc-nd/4.0)

关键词： Authorship attribution Data representation Pattern recognition Stylometry Supervised discretisation Unsupervised discretisation

来源：评论

学校读者我要写书评

暂无评论

Optimizing Artificial Neural Networks Trough Weight Adjustments

引用

Procedia computer science 2024年 246卷 2158-2165页

作者： Syed Muhammad Abrar Akber Agnieszka Szczesna Sadia Nishat Kazmi Department of Computer Graphics Vision and Digital Systems Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Gliwice 44-100 Poland Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Gliwice 44-100 Poland

Longer training times pose a significant challenge in Artificial neural networks (ANNs) as it may leads to increasing the computational costs and decreasing the effectiveness of the model. Therefore, it is imperative to reduce training times in ANNs to enhance the computational efficiency. The initialization of the weights between the layers in ANN plays a vital role in reducing training times. Appropriate weight initialization can help the network converge faster during the training by providing an optimum starting point for the network. Therefore, weight initialization techniques are essential for efficient training of ANNs. This paper revisits and implements different popular weight initialization techniques in ANNs and analyzes their impact on training time. Specifically, this paper implements Gaussian-based, Kaming-based, and Xavier-based weight initiation atop a popular DNN-based network. The experiments are conducted by employing a well-known dataset. The results show that the scenario when no weight initiation is applied consumed the highest training time, whereas different weight initiation techniques contribute in reducing the training times for the network.

关键词： Artificial neural network Deep neural network weight initiation

来源：评论

学校读者我要写书评

暂无评论

Attack Stealthiness and Detection of Multiagent Systems: A Zero-Sum Formulation 63

Attack Stealthiness and Detection of Multiagent Systems: A Z...

引用

63rd IEEE Conference on Decision and control, CDC 2024

作者： Zhu, Shiyong Wang, Miaomiao Chen, Jie Rantzer, Anders City University of Hong Kong Department of Electrical Engineering Hong Kong Hong Kong University of Science and Technology Department of Electronic and Computer Engineering Hong Kong Lund University Department of Automatic Control Lund Sweden

ISBN: (纸本)9798350316339

Cybersecurity has in recent years emerged as a paramount concern in the design and operation of industrial systems and civil infrastructures, due mainly to their susceptibility to malicious cyber attacks which take advantage of the vulnerability of communication networks and IT devices. In this paper, we investigate such an attack and counter attack scenario by considering multiagent systems, a somewhat basic prototype of cyberphysical systems. We study false data injection attacks launched on the agent sensors, and possible defense of such attacks at the agent actuators. The primary issue under consideration is the stealthiness of the attacks, while steering a multiagent system away from its consensual state. We propose a metric to quantify the stealthiness, and formulate the stealthiness problem as one of zero-sum games. We solve the problem explicitly, which gives rise to a fundamental bound on the stealthiness achievable, and as well optimal attack and defense strategies that achieve the optimal stealthiness, both of which can be obtained in terms of certain augmented controllability Gramians associated with the agents. The stealthiness bound is seen to depend on agent dynamics and network characteristics including a measure of connectivity. © 2024 IEEE.

关键词： Cyber attacks

来源：评论

学校读者我要写书评

暂无评论

Ante-Hoc Generation of Task-Agnostic Interpretation Maps

Ante-Hoc Generation of Task-Agnostic Interpretation Maps

引用

2023 IEEE/CVF Conference on computer Vision and Pattern Recognition Workshops, CVPRW 2023

作者： Akash Guna, R.T. Benitez, Raul Sikha, O.K. Amrita School of Engineering Department of Computer Science and Engineering Coimbatore India Universitat Politècnica de Catalunya BarcelonaTech Departament of Automatic Control Spain

ISBN: (纸本)9798350302493

Existing explainability approaches for convolutional neural networks (CNNs) are mainly applied after training (post-hoc) which is generally unreliable. Ante-hoc explainers trained simultaneously with the CNN are more reliable. However, current ante-hoc explanation methods mainly generate inexplicit concept-based explanations tailored to specific tasks. To address these limitations, we propose a task-agnostic ante-hoc framework that can generate interpretation maps to visually explain any CNN. Our framework simultaneously trains the CNN and a weighting network - an explanation generation module. The generated maps are self-explanatory, eliminating the need for manual identification of concepts. We demonstrate that our method can interpret classification, facial landmark detection, and image captioning tasks. We show that our framework is explicit, faithful, and stable through experiments. To the best of our knowledge, this is the first ante-hoc CNN explanation strategy that produces visual explanations generic to CNNs for different tasks. © 2023 IEEE.

关键词： computer vision

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：