检索结果-内蒙古大学图书馆

Fast DCT-I, DCT-III, and DCT-IV via Moments

Eurasip Journal on Applied Signal Processing 2005年第12期2005卷 1902-1909页

作者： Liu, J.G. Liu, Y.Z. Wang, G.Y. Key Laboratory of Education Huazhong University of Science and Technology Ministry for Image Processing and Intelligence Control Wuhan 430074 China Department of Computer Science and Engineering China University of Geosciences Wuhan 430074 China Key Laboratory of State Education Ministry for Image Processing and Intelligent Control Huazhong University of Science and Technology China Institute of Pattern Recognition and Artificial Intelligence Huazhong University of Science and Technology China

This paper presents a novel approach to compute DCT-I, DCT-III, and DCT-IV. By using a modular mapping and truncating, DCTs are approximated by linear sums of discrete moments computed fast only through additions. This enables us to use computational techniques developed for computing moments to compute DCTs efficiently. We demonstrate this by applying our earlier systolic solution to this problem. The method can also be applied to multidimensional DCTs as well as their inverses. © 2005 Hindawi Publishing Corporation.

关键词： Method of moments

来源：评论

学校读者我要写书评

暂无评论

UNIFORMER: UNIFIED TRANSFORMER FOR EFFICIENT SPATIOTEMPORAL REPRESENTATION LEARNING 10

UNIFORMER: UNIFIED TRANSFORMER FOR EFFICIENT SPATIOTEMPORAL ...

引用

10th International Conference on Learning Representations, ICLR 2022

作者： Li, Kunchang Wang, Yali Gao, Peng Song, Guanglu Liu, Yu Li, Hongsheng Qiao, Yu ShenZhen Key Lab of Computer Vision and Pattern Recognition SIAT-SenseTime Joint Lab Shenzhen Institute of Advanced Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences China Shanghai AI Laboratory Shanghai China SenseTime Research The Chinese University of Hong Kong Hong Kong

It is a challenging task to learn rich and multi-scale spatiotemporal semantics from high-dimensional videos, due to large local redundancy and complex global dependency between video frames. The recent advances in this research have been mainly driven by 3D convolutional neural networks and vision transformers. Although 3D convolution can efficiently aggregate local context to suppress local redundancy from a small 3D neighborhood, it lacks the capability to capture global dependency because of the limited receptive field. Alternatively, vision transformers can effectively capture long-range dependency by self-attention mechanism, while having the limitation on reducing local redundancy with blind similarity comparison among all the tokens in each layer. Based on these observations, we propose a novel Unified transFormer (UniFormer) which seamlessly integrates merits of 3D convolution and spatiotemporal self-attention in a concise transformer format, and achieves a preferable balance between computation and accuracy. Different from traditional transformers, our relation aggregator can tackle both spatiotemporal redundancy and dependency, by learning local and global token affinity respectively in shallow and deep layers. We conduct extensive experiments on the popular video benchmarks, e.g., Kinetics-400, Kinetics-600, and Something-Something V1&V2. With only ImageNet-1K pretraining, our UniFormer achieves 82.9%/84.8% top-1 accuracy on Kinetics-400/Kinetics-600, while requiring 10× fewer GFLOPs than other state-of-the-art methods. For Something-Something V1 and V2, our UniFormer achieves new state-of-the-art performances of 60.9% and 71.2% top-1 accuracy respectively. Code is available at https://***/Sense-X/UniFormer. © 2022 ICLR 2022 - 10th International Conference on Learning Representationss. All rights reserved.

关键词： Redundancy

来源：评论

学校读者我要写书评

暂无评论

Robust visual tracking based on online learning of joint sparse dictionary

Robust visual tracking based on online learning of joint spa...

引用

6th International Conference on Machine vision, ICMV 2013

作者： Li, Qiaozhe Qiao, Yu Yang, Jie Bai, Li Key Laboratory of Ministry of Education for System Control and Information Processing Institute of Image Processing and Pattern Recognition Shanghai Jiaotong University Shanghai China School of Computer Science University of Nottingham Nottingham United Kingdom

ISBN: (纸本)9780819499967

In this paper, we propose a robust visual tracking algorithm based on online learning of a joint sparse dictionary. The joint sparse dictionary consists of positive and negative sub-dictionaries, which model foreground and background objects respectively. An online dictionary learning method is developed to update the joint sparse dictionary by selecting both positive and negative bases from bags of positive and negative image patches/templates during tracking. A linear classifier is trained with sparse coefficients of image patches in the current frame, which are calculated using the joint sparse dictionary. This classifier is then used to locate the target in the next frame. Experimental results show that our tracking method is robust against object variation, occlusion and illumination change. © 2013 SPIE.

关键词： E-learning

来源：评论

学校读者我要写书评

暂无评论

Artificial intelligent techniques and its applications

Artificial intelligent techniques and its applications

引用

作者： Sundhararajan, Mahalingam Gao, Xiao-Zhi Vahdat Nejad, Hamed Department of Electronics and Communication Engineering Bharath Institute of Higher Education and Research Bharath University Chennai India Department of Machine Vision and Pattern Recognition Laboratory Lappeenranta University of Technology Finland Department of Computer Science and Engineering University of Birjand Birjand Iran

This special issue of the Journal of Intelligent & Fuzzy Systems is a selected collection of papers submitted to the IEEE International Conference on Algorithms, methodology, models and applications in emerging technologies and International Conference on Telecommunication, Power analysis and Computing Techniques and held from February 16-18, 2017 and April 6-8, 2017, Chennai, India. These papers have been reviewed and accepted for presentation at the conference and for publication in the Journal of Intelligent & Fuzzy Systems (JIFS). In this special issue there are 50 papers covering a wide range of tools, techniques and applications of artificial intelligent techniques and applications.

关键词： Fuzzy systems

来源：评论

学校读者我要写书评

暂无评论

Motion constraint patterns

Motion constraint patterns

引用

1993 IEEE Workshop on Qualitative vision, WQV 1993

作者： Fermüller, Cornelia Computer Vision Laboratory Center for Automation Research University of Maryland College ParkMD20742-3275 United States Department for Pattern Recognition and Image Processing Institute for Automation Technical University Vienna Treitlstraße 3 ViennaA-1040 Austria

ISBN: (纸本)0818636920

The problem of egomotion recovery has been treated by using as input local image motion, with the published algorithms utilizing the geometric constraint relating 2-D local image motion (optical flow, correspondence, derivatives of the image flow) to 3-D motion and structure. Since it has proved very difficult to achieve accurate input (local image motion), a lot of effort has been devoted to the development of robust techniques. A new approach to the problem of egomotion estimation is taken, based on constraints of a global nature. It is proved that local normal flow measurements form global patterns in the image plane. The position of these patterns is related to the three dimensional motion parameters. By locating some of these patterns, which depend only on subsets of the motion parameters, through a simple search technique, the 3-D motion parameters can be found. The proposed algorithmic procedure is very robust, since it is not affected by small perturbations in the normal flow measurements. As a matter of fact, since only the sign of the normal flow measurement is employed, the direction of translation and the axis of rotation can be estimated with up to 100% error in the image measurements. © 1993 IEEE.

关键词： Flow measurement

来源：评论

学校读者我要写书评

暂无评论

Advances in Biometric Person Authentication 1

引用

丛书名： Lecture Notes in computer Science

1000年

作者： Stan Z. Li Zhenan Sun Tieniu Tan Sharath Pankanti Gérard Chollet David Zhang

来源：评论

学校读者我要写书评

暂无评论

Dlung:Unsupervised Few-Shot Diffeomorphic Respiratory Motion Modeling

引用

Journal of Shanghai Jiaotong university(Science) 2023年第4期28卷 536-545页

作者：陈培芝郭逸凡王大寒陈金铃 College of Computer and Information Engineering Xiamen University of TechnologyXiamen 361024FujianChina Fujian Key Laboratory of Pattern Recognition and Image Understanding Xiamen 361024FujianChina School of Information Engineering Changchun Sci-Tech UniversityChangchun 130600China Department of Computer Science and Information Engineering Chaoyang University of TechnologyTaichung 41349TaiwanChina

Lung image registration plays an important role in lung analysis applications,such as respiratory motion *** learning-based image registration methods that can compute the deformation without the requirement of supervision attract much ***,it is noteworthy that they have two drawbacks:they do not handle the problem of limited data and do not guarantee diffeomorphic(topologypreserving)properties,especially when large deformation exists in lung *** this paper,we present an unsupervised few-shot learning-based diffeomorphic lung image registration,namely *** employ fine-tuning techniques to solve the problem of limited data and apply the scaling and squaring method to accomplish the diffeomorphic ***,atlas-based registration on spatio-temporal(4D)images is performed and thoroughly compared with baseline *** achieves the highest accuracy with diffeomorphic *** constructs accurate and fast respiratory motion models with limited *** research extends our knowledge of respiratory motion modeling.

关键词： unsupervised few-shot learning respiratory motion modeling diffeomorphic registration

来源：评论

学校读者我要写书评

暂无评论

Answer-supervised question reformulation for enhancing conversational machine comprehension 2

Answer-supervised question reformulation for enhancing conve...

引用

2nd Workshop on Machine Reading for Question Answering, MRQA@EMNLP 2019

作者： Li, Qian Su, Hui Niu, Cheng Wang, Daling Li, Zekang Feng, Shi Zhang, Yifei School of Computer Science and Engineering Northeastern University Shenyang China Pattern Recognition Center WeChat Ai Tencent Inc China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences

ISBN: (纸本)9781950737819

In conversational machine comprehension, it has become one of the research hotspots integrating conversational history information through question reformulation for obtaining better answers. However, the existing question reformulation models are trained only using supervised question labels annotated by annotators without considering any feedback information from answers. In this paper, we propose a novel Answer-Supervised Question Reformulation (ASQR) model for enhancing conversational machine comprehension with reinforcement learning technology. ASQR utilizes a pointer-copy-based question reformulation model as an agent, takes an action to predict the next word, and observes a reward for the whole sentence state after generating the end-of-sequence token. The experimental results on QuAC dataset prove that our ASQR model is more effective in conversational machine comprehension. Moreover, pretraining is essential in reinforcement learning models, so we provide a high-quality annotated dataset for question reformulation by sampling a part of QuAC dataset. © 2019 MRQA@EMNLP 2019 - Proceedings of the 2nd Workshop on Machine Reading for Question Answering. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Multi-Task Learning with Deep Dual-Path Network for Facial Attribute recognition 2020

Multi-Task Learning with Deep Dual-Path Network for Facial A...

引用

Proceedings of the 2020 9th International Conference on Computing and pattern recognition

作者： Xinyu Lai Si Chen Da-Han Wang Shunzhi Zhu School of Computer and Information Engineering Xiamen University of Technology Fujian Key Laboratory of Pattern Recognition and Image Understanding School of Computer and Information Engineering Xiamen University of Technology Fujian Key Laboratory of Pattern Recognition and Image Understanding

ISBN: (纸本)9781450387835

Facial attribute recognition is a popular and challenging research topic in computer vision. In the traditional deep learning based attribute recognition methods, the mid-level network features and the differences between attribute groups are not fully explored. To solve the above problem, a deep dual-path network is proposed for facial attribute recognition. In the multi-task learning framework, two sub-networks are employed to respectively extract the features of two attribute groups, i.e., local attributes and global ones, and designed with both different scale images and different depth networks. Furthermore, an adaptive Focal loss penalty scheme is developed to automatically assign weights to handle the class imbalance problem for facial attribute recognition. Experimental results on the challenging CelebA dataset show that the proposed method achieves the better performance than state-of-the-art methods.

关键词： multi-task learning dual-path network Facial attribute recognition attribute groups

来源：评论

学校读者我要写书评

暂无评论

A New Method for Detecting Altered Text in Document Images 2nd

A New Method for Detecting Altered Text in Document Images

引用

2nd International Conference on pattern recognition and Artificial Intelligence, ICPRAI 2020

作者： Nandanwar, Lokesh Shivakumara, Palaiahnakote Pal, Umapada Lu, Tong Lopresti, Daniel Seraogi, Bhagesh Chaudhuri, Bidyut B. Faculty of Computer Science and Information Technology University of Malaya Kuala Lumpur Malaysia Computer Vision and Pattern Recognition Unit Indian Statistical Institute Kolkata India National Key Lab for Novel Software Technology Nanjing University Nanjing China Computer Science and Engineering Lehigh University BethlehemPA United States

ISBN: (纸本)9783030598297

As more and more office documents are captured, stored, and shared in digital format, and as image editing software becomes increasingly more powerful, there is a growing concern about document authenticity. For example, texts in property documents can be altered to make an illegal deal, or the date on an airline ticket can be altered to gain entry to airport terminals by breaching security. To prevent such illicit activities, this paper presents a new method for detecting altered text in a document. The proposed method explores the relationship between positive and negative coefficients of a DCT to extract the effect of distortions caused by tampering operations. Here we divide DCT coefficients into positive and negative classes, then reconstructs images from the inverse DCT of the respective positive and negative coefficients. Next, we perform Laplacian filtering over reconstructed images for widening the gap between the values of text and other pixels. Then filtered images of positive and negative coefficients are fused by an average operation. For a fused image, we generate Canny and Sobel edge images in order to investigate the effect of distortion through quality measures, namely, MSE, PSNR and SSIM used as features. In addition, for the fused image, the proposed method extracts features based on histograms over the residual images. The features are then passed on to a deep Convolutional Neural Network for classification. The proposed method is tested on our own dataset as well as two standard datasets, namely IMEI and the ICPR 2018 Fraud Contest dataset. The results show that the proposed method is effective and outperforms existing methods. © 2020, Springer Nature Switzerland AG.

关键词： Inverse problems

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：