检索结果-内蒙古大学图书馆

IEEE International Conference on Robotics and Biomimetics

作者： Wang Huan Huiwen Guo Xinyu Wu Shenzhen Key Lab for Computer Vision and Pattern Recognition University of Chinese Academy of Sicences Department of Mechanical and Automation Engineering The Chinese University of Hong Kong

ISBN: (纸本)9781479973989

Most existing methods for abnormal event detection in the literature are relied on a training phase. Different from conventional approaches for abnormal event detection, a saliency attention based abnormal event detection approach is proposed in this paper. It is inspired by the visual attention mechanism that abnormal events are those which attract attention mostly in videos. The temporal and spatial abnormal saliency maps are firstly constructed and then the final abnormal event map is formatted by fusing them using a method with dynamic coefficients. The temporal abnormal saliency map is constructed by motion contrast between keypoints extracted from two successive video frames. The spatial abnormal saliency map is structured based on the color contrasts. Experiments performed on the benchmark datasets show that the proposed method achieves a high accurate and robust results for abnormal event detection without a training phase.

关键词： Clustering algorithms Hidden Markov models Event detection Conferences Color computer vision pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Context-transformer: Tackling object confusion for few-shot detection

arXiv

引用

arXiv 2020年

作者： Yang, Ze Wang, Yali Chen, Xianyu Liu, Jianzhuang Qiao, Yu ShenZhen Key Lab of Computer Vision and Pattern Recognition SIAT-SenseTime Joint Lab. Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Huawei Noah’s Ark Lab. SIAT Branch Shenzhen Institute of Artificial Intelligence and Robotics for Society

Few-shot object detection is a challenging but realistic scenario, where only a few annotated training images are availab.e for training detectors. A popular approach to handle this problem is transfer learning, i.e., fine-tuning a detector pretrained on a source-domain benchmark. However, such transferred detector often fails to recognize new objects in the target domain, due to low data diversity of training samples. To tackle this problem, we propose a novel Context-Transformer within a concise deep transfer framework. Specifically, Context-Transformer can effectively leverage source-domain object knowledge as guidance, and automatically exploit contexts from only a few training images in the target domain. Subsequently, it can adaptively integrate these relational clues to enhance the discriminative power of detector, in order to reduce object confusion in few-shot scenarios. Moreover, Context-Transformer is flexibly embedded in the popular SSD-style detectors, which makes it a plug-and-play module for end-to-end few-shot learning. Finally, we evaluate Context-Transformer on the challenging settings of few-shot detection and incremental few-shot detection. The experimental results show that, our framework outperforms the recent state-of-the-art approaches. The codes are availab.e at https://***/Ze-Yang/ContextTransformer. Copyright © 2020, The Authors. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Multi-modal pipeline for comprehensive validation of Mitral Valve geometry and functional computational models

Multi-modal pipeline for comprehensive validation of Mitral ...

引用

4th International Workshop on Statistical Atlases and Computational Models of the Heart: Imaging and Modelling Challenges, STACOM 2013, Held in Conjunction with MICCAI 2013

作者： Neumann, Dominik Grbic, Sasa Mansi, Tommaso Voigt, Ingmar Rabbah, Jean-Pierre Siefert, Andrew W. Saikrishnan, Neelakantan Yoganathan, Ajit P. Yuh, David D. Ionasec, Razvan Ioan Imaging and Computer Vision Siemens Corporate Technology Princeton NJ United States Pattern Recognition Lab. University of Erlangen-Nuremberg Germany Computer Aided Medical Procedures Technical University Munich Germany Wallace H. Coulter Department of Biomedical Engineering Georgia Institute of Technology Emory University Atlanta GA United States Section of Cardiac Surgery Department of Surgery Yale University New Haven CT United States

ISBN: (纸本)9783642542671

Valvular heart disease affects a high number of patients, exhibiting significant mortality and morbidity rates. Mitral Valve (MV) Regurgitation, a disorder in which the MV does not close properly during systole, is among its most common forms. Traditionally, it has been treated with MV replacement. However, recently there is an increased interest in MV repair procedures, providing better long-term survival, better preservation of heart function, lower risk of complications, and usually eliminating the need for long-term use of blood thinners (anticoagulants). These procedures are complex and require an experienced surgeon and elab.rate pre-operative planning. Hence, there is a need for efficient tools for training and planning of MV repair interventions. Computational models of valve function have been developed for these purposes. Nevertheless, state-of-the-art models remain approximations of real anatomy with considerable simplifications, since current modalities are limited by image quality. Hence, there is an important need to validate such low-fidelity models against comprehensive ex-vivo data to assess their clinical applicability. As a first step towards this aim, we propose an integrated pipeline for the validation of MV geometry and function models estimated in ex-vivo TEE data with respect to ex-vivo microCT data. We utilize a controlled experimental setup for ex-vivo imaging and employ robust machine learning and optimization techniques to extract reproducible geometrical models from both modalities. Using one exemplary case, we demonstrate the validity of our framework. © Springer-Verlag Berlin Heidelberg 2014.

关键词： Pipelines

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Feature Learning for Writer Identification and Writer Retrieval

arXiv

引用

arXiv 2017年

作者： Christlein, Vincent Gropp, Martin Fiel, Stefan Maier, Andreas Pattern Recognition Lab Friedrich-Alexander-Universität Erlangen-Nürnberg Erlangen91058 Germany Computer Vision Lab TU Wien Vienna1040 Austria

Deep Convolutional Neural Networks (CNN) have shown great success in supervised classification tasks such as character classification or dating. Deep learning methods typically need a lot of annotated training data, which is not availab.e in many scenarios. In these cases, traditional methods are often better than or equivalent to deep learning methods. In this paper, we propose a simple, yet effective, way to learn CNN activation features in an unsupervised manner. Therefore, we train a deep residual network using surrogate classes. The surrogate classes are created by clustering the training dataset, where each cluster index represents one surrogate class. The activations from the penultimate CNN layer serve as features for subsequent classification tasks. We evaluate the feature representations on two publicly availab.e datasets. The focus lies on the ICDAR17 competition dataset on historical document writer identification (Historical-WI). We show that the activation features trained without supervision are superior to descriptors of state-of-the-art writer identification methods. Additionally, we achieve comparable results in the case of handwriting classification using the ICFHR16 competition dataset on historical Latin script types (CLaMM16). Copyright © 2017, The Authors. All rights reserved.

关键词： Chemical activation

来源：评论

学校读者我要写书评

暂无评论

Color image filtering based on impulsive noise detection

Color image filtering based on impulsive noise detection

引用

MIPPR 2007: Automatic Target recognition and Image Analysis;and Multispectral Image Acquisition

作者： Li, Dehua Jin, Lianghai Yao, Xingzhong Xiong, Caiquan State Key Lab. for Image Processing and Intelligent Control Institute for Pattern Recognition and Artificial Intelligence Huazhong University of Science and Technology Wuhan 430074 China Second Artillery Command College Wuhan 430012 China School of Computer Science Hubei University of Technology Wuhan 430068 China

ISBN: (纸本)9780819469502

This paper introduces a new class of switching vector median filter. The proposed algorithm first uses four directional masks to analyze the color difference between the central pixel and its neighborhood pixels in the RGB color space and classify each color pixel into noisy pixel or noise-free one, and then employs the standard vector median filtering operations in the detected noisy locations to restore the corrupted pixels and leave the noise-free ones unchanged. The simulation results show that the proposed method excellently suppresses impulsive noise as well as preserving the image details well, and significantly outperforms the existing vector filtering solutions in terms of both the objective measures and the perceptual visual quality.

关键词： Color image processing

来源：评论

学校读者我要写书评

暂无评论

Classification of gothic and baroque architectural elements

Classification of gothic and baroque architectural elements

引用

International Conference on Systems, Signals and Image Processing, IWSSIP

作者： Gayane Shalunts Yll Haxhimusa Robert Sablatnig Institute of Computer Aided Automation Computer Vision Laboratory University of Technology Vienna Austria Institute of Computer Graphics and Algorithms Pattern Recognition and Image Processing Lab University of Technology Vienna Austria

Architectural elements are the components and details of buildings. Their unique set, combination, design, construction technique form the architectural style of buildings. Building facade classification by architectural styles is viewed as a task of classifying separate architectural structural elements. In the scope of building facade architectural style classification the current paper targets the problem of classification of Gothic and Baroque architectural elements called tracery, pediment and balustrade. Since certain gradient directions dominate on the shape of each architectural element, discrimination between dominating gradients means classification of architectural elements and thus architectural styles. We use local features to describe gradient directions. Our approach is based on clustering and learning of local features and yields a high classification rate.

关键词： Shape Three dimensional displays Visualization computer vision Histograms Windows

来源：评论

学校读者我要写书评

暂无评论

Application of PR techniques to mail sorting in China

Application of PR techniques to mail sorting in China

引用

2014 7th International C Conference on computer Science and Software Engineering, C3S2E 2014

作者： Liu, Li Lu, Shujing Lu, Yue Suen, Ching Y. Department of Computer Science and Technology East China Normal University Shanghai 200241 China ECNU-SRI Joint Lab. for Pattern Analysis and Intelligent System Shanghai Research Institute of China Post Group Shanghai 200062 China Centre for Pattern Recognition and Machine Intelligence Concordia University Montreal QC H3G 1M8 Canada

ISBN: (纸本)9781450327121

Mail sorting machines play an important role in postal automation. In this paper, we give a brief overview of mail sorting machines in China Post from a pattern recognition point of view. OCR techniques such as postcode recognition and address recognition are essential for mail sorting machines, which are considered as class imbalance problems in our study. In addition to OCR that mainly focuses on the characters in the image, some image-level features are exploited for near-duplicate envelope image matching, which is helpful in tracking pieces of mail in the network environment. © 2014 ACM.

关键词： Image matching

来源：评论

学校读者我要写书评

暂无评论

Deformation Robust Text Spotting with Geometric Prior

arXiv

引用

arXiv 2023年

作者： Hao, Xixuan Zhang, Aozhong Meng, Xianze Fu, Bin ShenZhen Key Lab of Computer Vision and Pattern Recognition Shenzhen Institute of Advanced Technology Chinese Academy of Sciences China The University of Hong Kong Hong Kong

The goal of text spotting is to perform text detection and recognition in an end-to-end manner. Although the diversity of luminosity and orientation in scene texts has been widely studied, the font diversity and shape variance of the same character are ignored in recent works, since most characters in natural images are rendered in standard fonts. To solve this problem, we present a Chinese Artistic Dataset, termed as ARText, which contains 33, 000 artistic images with rich shape deformation and font diversity. Based on this database, we develop a deformation robust text spotting method (DR TextSpotter) to solve the recognition problem of complex deformation of characters in different fonts. Specifically, we propose a geometric prior module to highlight the important features based on the unsupervised landmark detection sub-network. A graph convolution network is further constructed to fuse the character features and landmark features, and then performs semantic reasoning to enhance the discrimination for different characters. The experiments are conducted on ARText and IC19-ReCTS datasets. Our results demonstrate the effectiveness of our proposed method. Copyright © 2023, The Authors. All rights reserved.

关键词： Character recognition

来源：评论

学校读者我要写书评

暂无评论

Orientation Robust Scene Text recognition in Natural Scene*

Orientation Robust Scene Text Recognition in Natural Scene*

引用

IEEE International Conference on Robotics and Biomimetics

作者： Xiaolong Chen Zhengfu Zhang Yu Qiao Jiangyu Lai Jian Jiang Zeyu Zhang Bin Fu Guangzhou Power Supply Bureau Co. Ltd. Guangzhou China ShenZhen Key Lab of Computer Vision and Pattern Recognition Chinese Academy of Sciences

In recent years, scene text recognition has achieved significant improvement and various state-of-the-art recognition approaches have been proposed. This paper focused on recognizing text in natural photos of equipment nameplates, which has wide applications in industrial automations. This task only receives little attentions in previous works. The challenge of this problem comes from multi-orientation, curved, noisy and blurry text patches in equipment nameplates. To address this problem, we propose a deep model for text recognition in multi-oriented nameplates, namely, Orientation Robust Scene Text recognition (ORSTR). Specifically, our model employs a rectification module to transform curved, distorted or multi-orientation text to near-horizontal text with a carefully designed rectification module. Once the near-horizontal text has been generated, recognition network will output the predictions of text patches. Our scene text recognition model achieves 90 . 8% recognition accuracy on equipment nameplate dataset which outperforms previous scene text recognition model (CRNN) about 0 . 8%. Several extensive experiments have been conducted to verify the effectiveness of our model.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A semantic model for video based face recognition

A semantic model for video based face recognition

引用

International Conference on Information and Automation (ICIA)

作者： Dihong Gong Kai Zhu Zhifeng Li Yu Qiao Shenzhen Key Lab of Computer Vision and Pattern Recognition Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences The Chinese University of Hong Kong Hong Kong

ISBN: (纸本)9781479913329

Video-based face recognition has attracted a great deal of attention in recent years due to its wide applications. The challenge of video-based face recognition comes from several aspects. First, video data involves many frames, which increases data size and processing complexity. Second, key frames extracted from videos are usually of high intra-personal discrepancy due to variations in expressions, poses, and illuminations. In order to address these problems, we propose a novel semantic based subspace model to improve the performance of video based face recognition. The basic idea is to construct an appropriate low-dimensional subspace for each person, upon which a semantic model is built to classify the key frames of the person into specific class. After the semantic classification, the key frames belonging to the same classes, i.e. the same semantics, are used to train the linear classifiers for recognition. Extensive experiments on a large face video database (XM2VTS) clearly show that our approach obtains a significant performance improvement over the traditional approaches.

关键词： Face Training Face recognition Semantics Video sequences Principal component analysis Probes

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：