检索结果-南通市图书馆

Extending generalized unsupervised manifold alignment

Science China(Information Sciences) 2022年第7期65卷 139-156页

作者： Xiaoyi YIN Zhen CUI Hong CHANG Bingpeng MA shiguang shan Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS) Institute of Computing Technology Chinese Academy of Sciences CAS Center for Excellence in Brain Science and Intelligence Technology School of Computer Science and Engineering Nanjing University of Science and Technology University of Chinese Academy of Sciences

Building connections between different data sets is a fundamental task in machine learning and related application community. With proper manifold alignment, the correspondences between data sets will assist us with comprehensive study of data processes and analyses. Despite the several progresses in semi-supervised and unsupervised scenarios, potent manifold alignment methods in generalized and realistic circumstances remain in absence. Besides, theretofore unsupervised algorithms seldom prove themselves mathematically. In this paper, we devise an efficient method to properly solve the unsupervised manifold alignment problem and denominate it as extending generalized unsupervised manifold alignment(EGUMA)method. More specifically, an explicit relaxed integer programming method is adopted to solve the unsupervised manifold alignment problem, which reconciles three factors covering the updated local structure matching, the the feature comparability and geometric preservation. An additional effort is retained on extending the Frank Wolfe algorithm to tacking our optimization problem. Besides our previous endeavors we adopt a new strategy for neighborhood discovery in the manifolds. The main advantages over previous methods accommodate(1) simultaneous alignment and discovery of manifolds;(2) complete unsupervised learning structure without any prerequisite correspondence;(3) more concise local geometry for the embedding space;(4) efficient alternative optimization;(5) strict mathematical analysis on the convergence and efficiency issues. Experiments on real-world applications verify the high accuracy and efficiency of our proposed method.

关键词： unsupervised manifold alignment Unsupervised Manifold Alignment relaxed integer programming Relaxed Integer Programming Frank Wolfe algorithm Frank Wolfe Algorithm

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Polynomial stacked-attention network for nationality classification

引用

Frontiers of Computer Science 2022年第4期16卷 215-217页

作者： Kunyan Li Jie Zhang shiguang shan Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences(CAS) Institute of Computing TechnologyCASBeijing100190China University of Chinese Academy of Sciences Beijing100049China

1Introduction In recent years,nationality and region inference has atracted more and_more attentions.Contemporary anthropological researches indicate that human faces are influenced by genes,and other external factors,e.g-,geographical environment[1].This relevance provides possible solutions for geographical area inference by faces.However,it remains a challenging task due to the tiny differences on faces for various countries,which is even hard for human beings to tell.

关键词： Polynomial s. network

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Learning efficient text-to-image synthesis via interstage cross-sample similarity distillation

引用

Science China(Information Sciences) 2021年第2期64卷 19-30页

作者： Fengling MAO Bingpeng MA Hong CHANG shiguang shan Xilin CHEN School of Information Science and Technology ShanghaiTech University Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS) Institute of Computing TechnologyChinese Academy of Sciences University of Chinese Academy of Sciences CAS Center for Excellence in Brain Science and Intelligence Technology

For a given text, previous text-to-image synthesis methods commonly utilize a multistage generation model to produce images with high resolution in a coarse-to-fine manner. However, these methods ignore the interaction among stages, and they do not constrain the consistent cross-sample relations of images generated in different stages. These deficiencies result in inefficient generation and discrimination. In this study, we propose an interstage cross-sample similarity distillation model based on a generative adversarial network(GAN) for learning efficient text-to-image synthesis. To strengthen the interaction among stages,we achieve interstage knowledge distillation from the refined stage to the coarse stages with novel interstage cross-sample similarity distillation blocks. To enhance the constraint on the cross-sample relations of the images generated at different stages, we conduct cross-sample similarity distillation among the stages. Extensive experiments on the Oxford-102 and Caltech-UCSD Birds-200-2011(CUB) datasets show that our model generates visually pleasing images and achieves quantitatively comparable performance with state-of-the-art methods.

关键词： generative adversarialnetwork (GAN) text-to-image synthesis knowledge distillation

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Fusing magnitude and phase features with multiple face models for robust face recognition

引用

Frontiers of Computer Science 2018年第6期12卷 1173-1191页

作者： Yan LI shiguang shan Ruiping WANG Zhen CUI Xilin CHEN Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS) Institute of Computing Technology (ICT)CASBeijing100190China University of Chinese Academy of Sciences Beijing 100049China School of Computer Science and Engineering Nanjing University of Science and TechnologyNanjing 210094China

High accuracy face recognition is of great importance for a wide variety of real-world applications. Although significant progress has been made in the last decades, fully automatic face recognition systems have not yet approached the goal of surpassing the human vision system, even in controlled conditions. In this paper, we propose an approach for robust face recognition by fusing two complementary features: one is Gabor magnitude of multiple scales and orientations and the other is Fourier phase encoded by spatial pyramid based local phase quantization (SPLPQ). To reduce the high dimensionality of both features, block-wise fisher discriminant analysis (BFDA) is applied and further combined by score-level fusion. Moreover, inspired by the biological cognitive mechanism, multiple face models are exploited to further boost the robustness of the proposed approach. We evaluate the proposed approach on three challenging databases, i.e., FRGC ver2.0, LFW, and CFW-p, that address two face classification scenarios, i.e., verification and identification. Experimental results consistently exhibit the complementarity of the two features and the performance boost gained by the multiple face models. The proposed approach achieved approximately 96% verification rate when FAR was 0.1% on FRGC ver2.0 Exp.4, impressively surpassing all the best known results.

关键词： face recognition fisher discriminant analysis fusion Gabor magnitude feature multiple face models spatial pyramid based local phase quantization

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

VIPLFaceNet： an open source deep face recognition SDK

引用

Frontiers of Computer Science 2017年第2期11卷 208-218页

作者： Xin LIU Meina KAN Wanglong WU shiguang shan Xilin CHEN Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS) Institute of Computing Technology CAS Beijing 100190 China University of Chinese Academy of Sciences Beijing 100049 China

Robust face representation is imperative to highly accurate face recognition. In this work, we propose an open source face recognition method with deep representation named as VIPLFaceNet, which is a lO-layer deep convolu- tional neural network with seven convolutional layers and three fully-connected layers. Compared with the well-known AlexNet, our VIPLFaceNet takes only 20% training time and 60% testing time, but achieves 40% drop in error rate on the real-world face recognition benchmark LFW. Our VIPLFaceNet achieves 98.60% mean accuracy on LFW us- ing one single network. An open-source C＋＋ SDK based on VIPLFaceNet is released under BSD license. The SDK takes about 150ms to process one face image in a single thread on an i7 desktop CPU. VIPLFaceNet provides a state-of-the-art start point for both academic and industrial face recognition applications.

关键词： deep learning face recognition open source,VIPLFaceNet

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Locality-constrained framework for face alignment

引用

Frontiers of Computer Science 2019年第4期13卷 789-801页

作者： Jie ZHANG Xiaowei ZHAO Meina KAN shiguang shan Xiujuan CHAI Xilin CHEN Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences Institute of Computing Technology Chinese Academy of Sciences Beijing 100190 China University of Chinese Academy of Sciences Beijing 100049 China Alibaba Group Hangzhou 311121 China

Although the conventional active appearance model (AAM) has achieved some success for face alignment, it still suffers from the generalization problem when be applied to unseen subjects and images. To deal with the gem eralization problem of AAM, we first reformulate the original AAM as sparsity-regularized AAM, which can achieve more compact/better shape and appearance priors by selecting nearest neighbors as the bases of the shape and appearance model. To speed up the fitting procedure, the sparsity in sparsity?regularized AAM is approximated by using the locality (i.e., AT-nearest neighbor), and thus inducing the locality-constrained active appearance model (LC-AAM). The LC-AAM solves a constrained AAM-like fitting problem with the K-nearest neighbors as the bases of shape and appearance model. To alleviate the adverse influence of inaccurate AT-nearest neighbor results, the locality constraint is further embedded in the discriminative fitting method denoted as LC-DFM, which can find better K-nearest neighbor results by employing shape-indexed feature, and can also tolerate some inaccurate neighbors benefited from the regression model rather than the generative model in AAM. Extensive experiments on several datasets demonstrate that our methods outperform the state-of-the-arts in both detection accuracy and generalization ability.

关键词： locality-constrained AAM locality-constrained DFM face alignment sparsity-regularization

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

ProMIS:概率图采样图像增广驱动的弱监督物体检测方法

引用

中国图象图形学报 2023年第7期28卷 2037-2053页

作者：李笑颜阚美娜梁浩山世光中国科学院智能信息处理重点实验室北京100190 中国科学院大学计算机科学与技术学院北京100049 鹏城实验室深圳518055

目的弱监督物体检测是一种仅利用图像类别标签训练物体检测器的技术。近年来弱监督物体检测器的精度不断提高,但在如何提升检出物体的完整性、如何从多个同类物体中区分出单一个体的问题上仍面临极大挑战。围绕上述问题,提出了基于物体... 详细信息

目的弱监督物体检测是一种仅利用图像类别标签训练物体检测器的技术。近年来弱监督物体检测器的精度不断提高,但在如何提升检出物体的完整性、如何从多个同类物体中区分出单一个体的问题上仍面临极大挑战。围绕上述问题,提出了基于物体布局后验概率图进行多物体图像增广的弱监督物体检测方法ProMIS(probability-based multi-object image synthesis)。方法将检出物体存储到物体候选池,并将候选池中的物体插入到输入图像中,构造带有伪边界框标注的增广图像,进而利用增广后的图像训练弱监督物体检测器。该方法包含图像增广与弱监督物体检测两个相互作用的模块。图像增广模块将候选池中的物体插入一幅输入图像,该过程通过后验概率的估计与采样对插入物体的类别、位置和尺度进行约束,以保证增广图像的合理性;弱监督物体检测模块利用增广后的多物体图像、对应的类别标签、物体伪边界框标签训练物体检测器,并将原始输入图像上检到的高置信度物体储存到物体候选池中。训练过程中,为了避免过拟合,本文在基线算法的基础上增加一个并行的检测分支,即基于增广边界框的检测分支,该分支利用增广得到的伪边界框标注进行训练,原有基线算法的检测分支仍使用图像标签进行训练。测试时,本文方法仅使用基于增广边界框的检测分支产生检测结果。本文提出的增广策略和检测器的分支结构在不同弱监督物体检测器上均适用。结果在Pascal VOC(pattern analysis,statistical modeling and computational learning visual object classes)2007和Pascal VOC 2012数据集上,将该方法嵌入到多种现有的弱监督物体检测器中,平均精度均值(mean average precision,mAP)平均获得了2.9%和4.2%的提升。结论本文证明了采用弱监督物体检测伪边界框标签生成的增广图像包含丰富信息,能够辅助弱监督检测器学习物体部件、整体以及多物体簇之间的区别。

关键词：弱监督物体检测多物体数据增广图像融合概率图采样后验概率估计

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

面向大姿态人脸识别的正面化形变场学习

引用

中国图象图形学报 2022年第7期27卷 2171-2184页

作者：胡蓝青阚美娜山世光陈熙霖中国科学院计算技术研究所北京100190 中国科学院大学计算机科学与技术学院北京100049

目的人脸识别已经得到了广泛应用,但大姿态人脸识别问题仍未完美解决。已有方法或提取姿态鲁棒特征,或进行人脸姿态的正面化。其中主流的人脸正面化方法包括2D回归生成和3D模型形变建模,前者能够生成相对自然真实的人脸,但会引入额外的... 详细信息

目的人脸识别已经得到了广泛应用,但大姿态人脸识别问题仍未完美解决。已有方法或提取姿态鲁棒特征,或进行人脸姿态的正面化。其中主流的人脸正面化方法包括2D回归生成和3D模型形变建模,前者能够生成相对自然真实的人脸,但会引入额外的噪声导致图像信息的扭曲;后者能够保持原始的人脸结构信息,但生成过程是基于物理模型的,不够自然灵活。为此,结合2D和3D方法的优势,本文提出了基于由粗到细形变场的人脸正面化方法。方法该形变场由深度网络以2D回归方式学得,反映的是不同视角人脸图像像素之间的语义级对应关系,可以类3D的方式实现非正面人脸图像的正面化,因此该方法兼具了2D正面化方法的灵活性与3D正面化方法的保真性,且借鉴分步渐进的思路,本文提出了由粗到细的形变场学习框架,以获得更加准确鲁棒的形变场。结果本文采用大姿态人脸识别实验来验证本文方法的有效性,在MultiPIE(multi pose,illumination,expressions)、LFW(labeled faces in the wild)、CFP(celebrities in frontal-profile in the wild)、IJB-A(intelligence advanced research projects activity Janus benchmark-A)等4个数据集上均取得了比已有方法更高的人脸识别精度。结论本文提出的基于由粗到细的形变场学习的人脸正面化方法,综合了2D和3D人脸正面化方法的优点,使人脸正面化结果的学习更加灵活、准确,保持了更多有利于识别的身份信息。

关键词：大姿态人脸识别人脸正面化可学习形变场由粗到细学习全卷积网络

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

结构化稀疏线性判别分析

引用

计算机研究与发展 2014年第10期51卷 2295-2301页

作者：崔振山世光陈熙霖华侨大学计算机科学与技术学院福建厦门361021 中国科学院智能信息处理重点实验室(中国科学院计算技术研究所) 北京100190 中国科学院大学北京100049

在监督场景下线性判别分析(linear discriminant analysis,LDA)是一种非常有效的特征提取方法.然而,LDA在小样本情况下通常会出现过拟合现象,并且学习的投影变换难以给出人类认知上的解释.针对这些问题,特别是可解释性结构的发现,借助于... 详细信息

在监督场景下线性判别分析(linear discriminant analysis,LDA)是一种非常有效的特征提取方法.然而,LDA在小样本情况下通常会出现过拟合现象,并且学习的投影变换难以给出人类认知上的解释.针对这些问题,特别是可解释性结构的发现,借助于LDA的线性回归模型和结构化稀疏L2,1范数,提出了结构化稀疏线性判别分析(structured sparse LDA,SSLDA)方法.进一步,为了去除线性变换间的相关性,提出了正交化的SSLDA(orthogonalized SSLDA,OSSLDA),它能更加有效地学习到细致的结构信息.为了求解这2个模型,引入了一个半二次的优化算法,它在投影变换和新引入的辅助变量之间采用交替优化的思想.为了验证所提出的方法,在AR、扩展的YaleB和MultiPIE 3个人脸数据库上对比了LDA及其变种方法,实验表明了所提出方法的有效性以及可解释性.

关键词：线性判别分析正交化人脸识别最小二乘结构化稀疏

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

双向层级结构分类器及其在面部特征点定位中的应用

引用

高技术通讯 2010年第8期20卷 816-821页

作者：牛志恒山世光陈熙霖哈尔滨工业大学计算机科学与技术学院哈尔滨150001 中国科学院计算技术研究所中国科学院智能信息处理重点实验室北京100190

设计了一种新颖的双向层级结构分类器,并将其应用于面部特征点的定位。该分类器在训练过程中对正例样本和反例样本交替进行重采样,因而与采用传统的单向层级分类器的面部特征定位方法相比,采用这种新的双向层级结构分类器的方法具有以... 详细信息

设计了一种新颖的双向层级结构分类器,并将其应用于面部特征点的定位。该分类器在训练过程中对正例样本和反例样本交替进行重采样,因而与采用传统的单向层级分类器的面部特征定位方法相比,采用这种新的双向层级结构分类器的方法具有以下优点:可以应对大规模的数据集;可以处理存在复杂变化的正反例样本;而且无论是在训练过程还是在测试过程,其算法都能快速地过滤大量的'易分'样本,执行效率非常高。在两个公开测试数据库上的实验结果表明,采用双向层级结构分类器的方法可以实现准确、快速的特征点定位。

关键词：双向层级结构分类器设计正(反)例样本面部特征点定位

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

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：