检索结果-内蒙古大学图书馆

Orthogonal Transform-Driven Data Augmentation for Limited Gaussian-Tainted Dataset

ieee ACCESS 2024年 12卷 127272-127282页

作者： Won Yoon, Jung Jun Yook, Hyun Min Hong, Pyo Kyu Lee, Youn Kim, Tae Hyung Hongik Univ Dept Comp Engn Seoul 04066 South Korea

A large amount of data collected from sensors exhibits Gaussian noise characteristics, making denoising and related processing critical. However, data scarcity can lead to overfitting, posing challenges in training deep learning-based denoising methods. While various data augmentation methods have been proposed, they do not provide a means to augment original data to large-scale data while preserving the exact noise distribution. To address this, we introduce a novel data augmentation method for data with additive white Gaussian noise (AWGN). Our method is based on two main premises: first, orthogonal transforms preserve the probability distribution of AWGN;second, the signals we aim to recover generally exhibit smooth characteristics, unlike noise. Building on these premises, we propose adaptive smoothness-promoting orthogonal transforms for augmenting limited existing data. We evaluated the proposed method in Gaussian denoising tasks with limited data and confirmed that it achieves substantial improvement in deep learning model performance, comparable to those obtained with sufficient data.

关键词： Data augmentation Gaussian noise smoothness-promoting orthogonal transform Data augmentation smoothness-promoting orthogonal transform Gaussian noise REFERENCES [1] A. Buades B. Coll and J.-M. Morel "A non-local algorithm for image denoising "in proc. ieee comput. soc. conf. comput. Vis. pattern Recognit. (CVPR) vol. 2 Jun. 2005 pp. 60-65. [2] K. Dabov A. Foi V. Katkovnik and K. Egiazarian "image denoising by sparse 3-D transform-domain collaborative filtering "ieee ieee Trans. image process. vol. 16 no. 8 pp. 2080-2095 Aug. 2007. [3] M. Elad and M. Aharon "image denoising via sparse and redundant representations over learned dictionaries vol. 15 no. 12 pp. 3736-3745 Dec. 2006. [4] S. V. Mohd Sagheer and S. N. George "A review on medical image denoising algorithms "Biomed. Biomed. Signal process. Control vol. 61 Aug. 2020 Art. no. 102036 [1] A. Buades "ieee Trans. image process. "Biomed. Signal process. Control

来源：评论

学校读者我要写书评

暂无评论

Decoupling-and-Aggregating for image Exposure Correction

Decoupling-and-Aggregating for Image Exposure Correction

引用

ieee/CVF conference on computer Vision and pattern recognition (CVPR)

作者： Wang, Yang Peng, Long Li, Liang Cao, Yang Zha, Zheng-Jun Univ Sci & Technol China Hefei Peoples R China Chinese Acad Sci Inst Comput Tech Key Lab Intell Info Proc Beijing Peoples R China

ISBN: (纸本)9798350301298

The images captured under improper exposure conditions often suffer from contrast degradation and detail distortion. Contrast degradation will destroy the statistical properties of low-frequency components, while detail distortion will disturb the structural properties of high-frequency components, leading to the low-frequency and high-frequency components being mixed and inseparable. This will limit the statistical and structural modeling capacity for exposure correction. To address this issue, this paper proposes to decouple the contrast enhancement and detail restoration within each convolution process. It is based on the observation that, in the local regions covered by convolution kernels, the feature response of low-/high-frequency can be decoupled by addition/difference operation. To this end, we inject the addition/difference operation into the convolution process and devise a Contrast Aware (CA) unit and a Detail Aware (DA) unit to facilitate the statistical and structural regularities modeling. The proposed CA and DA can be plugged into existing CNN-based exposure correction networks to substitute the Traditional Convolution (TConv) to improve the performance. Furthermore, to maintain the computational costs of the network without changing, we aggregate two units into a single TConv kernel using structural re-parameterization. Evaluations of nine methods and five benchmark datasets demonstrate that our proposed method can comprehensively improve the performance of existing methods without introducing extra computational costs compared with the original networks. The codes will be publicly available.

关键词： computational imaging

来源：评论

学校读者我要写书评

暂无评论

Attribute Group Editing for Reliable Few-shot image Generation

Attribute Group Editing for Reliable Few-shot Image Generati...

引用

ieee/CVF conference on computer Vision and pattern recognition (CVPR)

作者： Ding, Guanqi Han, Xinzhe Wang, Shuhui Wu, Shuzhe Jin, Xin Tu, Dandan Huang, Qingming Univ Chinese Acad Sci Beijing Peoples R China Chinese Acad Sci Key Lab Intell Info Proc Inst Comput Tech Beijing Peoples R China Huawei Cloud EI Innovat Lab Shanghai Peoples R China Peng Cheng Lab Shenzhen Peoples R China

ISBN: (数字)9781665469463

ISBN: (纸本)9781665469463

Few-shot image generation is a challenging task even using the state-of-the-art Generative Adversarial Networks (GANs). Due to the unstable GAN training process and the limited training data, the generated images are often of low quality and low diversity. In this work, we propose a new "editing-based" method, i.e., Attribute Group Editing (AGE), for few-shot image generation. The basic assumption is that any image is a collection of attributes and the editing direction for a specific attribute is shared across all categories. AGE examines the internal representation learned in GANs and identifies semantically meaningful directions. Specifically, the class embedding, i.e., the mean vector of the latent codes from a specific category, is used to represent the category-relevant attributes, and the category-irrelevant attributes are learned globally by Sparse Dictionary Learning on the difference between the sample embedding and the class embedding. Given a GAN well trained on seen categories, diverse images of unseen categories can be synthesized through editing category-irrelevant attributes while keeping category-relevant attributes unchanged. Without re-training the GAN, AGE is capable of not only producing more realistic and diverse images for downstream visual applications with limited data but achieving controllable image editing with interpretable category-irrelevant directions. Code is available at https://***/UniBester/AGE.

关键词： Training Visualization Codes image synthesis Semantics Training data Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

DeeCap: Dynamic Early Exiting for Efficient image Captioning

DeeCap: Dynamic Early Exiting for Efficient Image Captioning

引用

ieee/CVF conference on computer Vision and pattern recognition (CVPR)

作者： Fei, Zhengcong Yan, Xu Wang, Shuhui Tian, Qi Chinese Acad Sci Inst Comput Tech Key Lab Intell Info Proc Beijing Peoples R China Univ Chinese Acad Sci Beijing Peoples R China Peng Cheng Lab Shenzhen Peoples R China Huawei Cloud & AI Beijing Peoples R China

ISBN: (数字)9781665469463

ISBN: (纸本)9781665469463

Both accuracy and efficiency are crucial for image captioning in real-world scenarios. Although Transformerbased models have gained significant improved captioning performance, their computational cost is very high. A feasible way to reduce the time complexity is to exit the prediction early in internal decoding layers without passing the entire model. However, it is not straightforward to devise early exiting into image captioning due to the following issues. On one hand, the representation in shallow layers lacks high-level semantic and sufficient cross-modal fusion information for accurate prediction. On the other hand, the exiting decisions made by internal classifiers are unreliable sometimes. To solve these issues, we propose DeeCap framework for efficient image captioning, which dynamically selects proper-sized decoding layers from a global perspective to exit early. The key to successful early exiting lies in the specially designed imitation learning mechanism, which predicts the deep layer activation with shallow layer features. By deliberately merging the imitation learning into the whole image captioning architecture, the imitated deep layer representation can mitigate the loss brought by the missing of actual deep layers when early exiting is undertaken, resulting in significant reduction in calculation cost with small sacrifice of accuracy. Experiments on the MS COCO and Flickr30k datasets demonstrate the DeeCap can achieve competitive performances with 4x speed-up.

关键词： Learning systems computational modeling Semantics Merging Predictive models Transformers Decoding

来源：评论

学校读者我要写书评

暂无评论

Few Shot Generative Model Adaption via Relaxed Spatial Structural Alignment

Few Shot Generative Model Adaption via Relaxed Spatial Struc...

引用

ieee/CVF conference on computer Vision and pattern recognition (CVPR)

作者： Xiao, Jiayu Li, Liang Wang, Chaofei Zha, Zheng-Jun Huang, Qingming Chinese Acad Sci Inst Comput Tech Key Lab Intell Info Proc Beijing Peoples R China Univ Chinese Acad Sci Beijing Peoples R China Tsinghua Univ Dept Automat Beijing Peoples R China Univ Sci & Technol China Hefei Peoples R China Peng Cheng Lab Shenzhen Peoples R China

ISBN: (数字)9781665469463

ISBN: (纸本)9781665469463

Training a generative adversarial network (GAN) with limited data has been a challenging task A feasible solution is to start with a GAN well-trained on a large scale source domain and adapt it to the target domain with a few samples, termed as few shot generative model adaption. However, existing methods are prone to model overfitting and collapse in extremely few shot setting (less than 10). To solve this problem, we propose a relaxed spatial structural alignment (RSSA) method to calibrate the target generative models during the adaption. We design a cross-domain spatial structural consistency loss comprising the self-correlation and disturbance correlation consistency loss. It helps align the spatial structural information between the synthesis image pairs of the source and target domains. To relax the cross-domain alignment, we compress the original latent space of generative models to a subspace. image pairs generated from the subspace are pulled closer. Qualitative and quantitative experiments show that our method consistently surpasses the state-of-the-art methods in few shot setting. Our source code: https://***/StevenShaw1999/RSSA.

关键词： Training Adaptation models computer vision image coding Correlation Codes computational modeling

来源：评论

学校读者我要写书评

暂无评论

Automatic recognition of Learning Resource Category in a Digital Library 21

Automatic Recognition of Learning Resource Category in a Dig...

引用

21st ACM/ieee Joint conference on Digital Libraries (JCDL)

作者： Banerjee, Soumya Sanyal, Debarshi Kumar Chattopadhyay, Samiran Bhowmick, Plaban Kumar Das, Partha Pratim IIT Kharagpur Kharagpur 721302 W Bengal India Indian Assoc Cultivat Sci Kolkata 700032 India Jadavpur Univ Kolkata 700106 India

ISBN: (纸本)9781665417709

Digital libraries generally need to process a large volume of diverse document types. The collection and tagging of metadata is a long, error-prone, workforce-consuming task. We are attempting to build an automatic metadata extractor for digital libraries. In this work, we present the Heterogeneous Learning Resources (HLR) dataset for document image classification. The individual learning resource is first decomposed into its constituent document images (sheets) which are then passed through an OCR tool to obtain the textual representation. The document image and its textual content are classified with state-of-the-art classifiers. Finally, the labels of the constituent document images are used to predict the label of the overall document.

关键词： deep learning transfer learning digital library document image classification

来源：评论

学校读者我要写书评

暂无评论

State-Relabeling Adversarial Active Learning

State-Relabeling Adversarial Active Learning

引用

ieee/CVF conference on computer Vision and pattern recognition (CVPR)

作者： Zhang, Beichen Li, Liang Yang, Shijie Wang, Shuhui Zha, Zheng-Jun Huang, Qingming Univ Chinese Acad Sci Beijing Peoples R China Chinese Acad Sci Inst Comput Tech Key Lab Intell Info Proc Beijing Peoples R China Univ Sci & Technol China Hefei Peoples R China Peng Cheng Lab Shenzhen Peoples R China

ISBN: (数字)9781728171685

ISBN: (纸本)9781728171685

Active learning is to design label-efficient algorithms by sampling the most representative samples to be labeled by an oracle. In this paper, we propose a state relabeling adversarial active learning model (SRAAL), that leverages both the annotation and the labeled/unlabeled state information for deriving the most informative unlabeled samples. The SRAAL consists of a representation generator and a state discriminator. The generator uses the complementary annotation information with traditional reconstruction information to generate the unified representation of samples, which embeds the semantic into the whole data representation. Then, we design an online uncertainty indicator in the discriminator, which endues unlabeled samples with different importance. As a result, we can select the most informative samples based on the discriminator's predicted state. We also design an algorithm to initialize the labeled pool, which makes subsequent sampling more efficient. The experiments conducted on various datasets show that our model outperforms the previous state-of-art active learning methods and our initially sampling algorithm achieves better performance.

关键词： Task analysis Uncertainty Generators Data models computational modeling image reconstruction Learning systems

来源：评论

学校读者我要写书评

暂无评论

Multiple Kernel Collaborative Representation Based Classification 13

Multiple Kernel Collaborative Representation Based Classific...

引用

13th ieee International conference on Signal processing (ICSP)

作者： Li, Ru Zhang, Qian Gao, Zhiming Liu, Bao-Di Wang, Yanjiang China Univ Petr East China Coll Informat & Control Engn Qingdao Peoples R China Univ Elect Sci & Technol China Sch Elect Engn Chengdu Sichuan Peoples R China

ISBN: (纸本)9781509013456

At present, collaborative representation based classification (CRC) is widely used in many pattern classification and recognition tasks. Meanwhile, spatial pyramid matching (SPM) method, which considers the spatial information in representing the image, is efficient for image classification. However, for SPM, the weights to evaluate the representation of different subregions are fixed. In this paper, we combine CRC and multiple kernel learning approach, propose the multiple kernel collaborative representation based classification (MKCRC) method, and apply it to image classification to learn the weights of different subregions in representing the image. Experimental results have obvious advantages than state-of-the-art methods in several benchmark datasets, such as Caltech101, 102flowers, Scene 15 and UIUC sports..

关键词： image classification Multiple kernel learning Spatial pyramid matching

来源：评论

学校读者我要写书评

暂无评论

Foreword

Proceedings of the 2016 International Conference on Image Pr...

引用

proceedings of the 2016 International conference on image processing, computer Vision, and pattern recognition, IPCV 2016 2016年

作者： Abd-Wahab, Mohd Helmy Al-Bakry, Abbas Al-Holou, Nizar Arabnia, Hamid R. Bhattacharya, Mahua Martinez Castillo, Juan Jose Daimi, Kevin Deligiannidis, Leonidas Djoudi, Lamia Atma Duong, Trung Eshaghian-Wilner, Mary Mehrnoosh Gravvanis, George A. Huang, Ruizhu Jandieri, George Kim, Byung-Gyu Kim, Tai-hoon Korovin, Iakov Lai, Guoming Lee, Hyo Jong Bin Mansor, Muhammad Naufal Marsh, Andrew Mostafaeipour, Ali Park, James J. Patil, Shashikant Ponalagusamy, R. Schaefer, Gerald Singh, Akash Solo, Ashu M.G. Swee, Sim Kok Thomas, Jaya Tinetti, Fernando G. Vladimir, Hahanov Wang, Shiuh-Jeng Yang, Mary Yoe, Hyun You, Jane Zhao, Wenbing Department of Computer Engineering University Tun Hussein Onn Malaysia Malaysia University of IT and Communications Baghdad Iraq Electrical and Computer Engineering Department IEEE/SEM-Computer Chapter University of Detroit Mercy DetroitMI United States University of Georgia United States ABV Indian Institute of Information Technology and Management MHRD Government of India India Acantelys Alan Turing Nikola Tesla Research Group and GIPEB Universidad Nacional Abierta Venezuela Computer Science and Software Engineering Programs Department of Mathematics Computer Science and Software Engineering University of Detroit Mercy DetroitMI United States Department of Computer Information Systems Wentworth Institute of Technology BostonMA United States Synchrone Technologies France Rutgers University State University of New Jersey New Jersey United States University of Southern California California United States Electrical Engineering University of California Los Angeles Los Angeles [UCLA CA United States Advanced Scientific Computing Applied Math and Applications Research Group Applied Mathematics and Numerical Computing and Department of ECE School of Engineering Democritus University of Thrace Xanthi Greece Texas Advanced Computing Center University of Texas AustinTX United States Georgian Technical University Tbilisi Georgia Institute of Cybernetics Georgian Academy of Science Georgia Multimedia Processing CommunicationsLab.[MPCL Department of Computer Science and Engineering College of Engineering SunMoon University Korea Republic of School of Information and Computing Science University of Tasmania Australia Southern Federal University Russia Computer Science and Technology Sun Yat-Sen University Guangzhou China Center for Advanced Image and Information Technology Division of Computer Science and Engineering Chonbuk National University Korea Republic of Faculty of Engineering Technology Kampus Uniciti Alam Universiti Malaysia Perlis UniMAP Malaysi

来源：评论

学校读者我要写书评

暂无评论

STATISTICAL pattern recognition FOR REAL-TIME image EDGE DETECTION ON FPGA 12

STATISTICAL PATTERN RECOGNITION FOR REAL-TIME IMAGE EDGE DET...

引用

12th ieee International conference on Signal processing (ICSP)

作者： Liu, Ziyan Qi, Jia Feng, Liang Feng, Li Guizhou Univ Coll Elect & Informat Guiyang 550025 Peoples R China State Grid Chongqing Elect Power Co Chongqing 400014 Peoples R China

ISBN: (纸本)9781479921867

image edge detection is a fundamental process in computer vision. image edges represent the major fraction of information in an image. Traditional edge-detection techniques focus on the gradient calculation method. In this paper, for the first time, the statistical pattern recognition method is used to detect the edge after the real-time image was processed via the median filtering method and implemented on FPGA. In comparison to the Sobel algorithm, the proposed method has superior anti-noise capability.

关键词： Statistical pattern recognition Median Filtering FPGA Real-Time image Edge Detection

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：