检索结果-内蒙古大学图书馆

Combining local and global: Rich and robust feature pooling for visual recognition

PATTERN RECOGNITION 2017年第0期62卷 225-235页

作者： Xiong, Wei Zhang, Lefei Du, Bo Tao, Dacheng Wuhan Univ Key Lab Aerosp Informat Secur & Trusted Comp Minist Educ State Key Lab Software EngnSch Comp Wuhan 430072 Hubei Peoples R China Univ Technol Ctr Quantum Computat & Intelligent Syst Sydney NSW 2007 Australia

The human visual system proves expert in discovering patterns in both global and local feature space. Can we design a similar way for unsupervised feature learning? In this paper, we propose a novel spatial pooling method within an unsupervised feature learning framework, named Rich and Robust Feature Pooling ((RFP)-F-2), to better extract rich and robust representation from sparse feature maps learned from the raw data. Both local and global pooling strategies are further considered to instantiate such a method. The former selects the most representative features in the sub-region and summarizes the joint distribution of the selected features, while the latter is utilized to extract multiple resolutions of features and fuse the features with a feature balance kernel for rich representation. Extensive experiments on several image recognition tasks demonstrate the superiority of the proposed method. (C) 2016 Elsevier Ltd. All rights reserved.

关键词： Unsupervised learning Spatial pooling auto-encoder

来源：评论

学校读者我要写书评

暂无评论

autoencoder-based Feature Learning from a 2D Depth Map and 3D Skeleton for Action Recognition

引用

电脑学刊 2018年第4期29卷 82-95页

作者： Zhi-Ze Wu Shou-Hong Wan Li Yan Li-Hua Yue

3D skeleton is compact for human action recognition. Existing approaches focus mainly on performing action recognition with only joint coordinates. However, they are difficult recognizing some similar or complex actions based on skeleton data only. Furthermore, they could have poor performance when the estimated skeletal joints are not reliable or when the actions share a large overlap in the sequences. Noticing that the existing 3D skeleton data is usually obtained from depth map and a neural network can simulate arbitrary relationships, we propose an efficient approach for action recognition by combining 2D depth information with 3D skeleton information using a new deep architecture called Deep Multimodal auto-encoder (DMAE) in this paper. First, the framework of DMAE employs two auto-encoders to extract hidden representations for the 2D depth maps and the 3D skeletons. Second, DMAE uses a two-layer neural network to map the hidden representations of the 2D depth maps to those of the 3D skeletons. Third, based on a Back-propagation Neural Network (BP-NN), it can jointly explore the hidden 2D/3D representations and the appropriate relationships between them. We finally use Temporal Pyramid Matching (TPM) on the learned features to generate temporal representations and perform classifications using a linear SVM. Additionally, we demonstrate the effectiveness and efficiency of DMAE by extensive experimentation using two popular action datasets.

关键词： 2D depth map 3D skeleton action recognition auto-encoder back-propagation neural network

来源：评论

学校读者我要写书评

暂无评论

HSAE: A Hessian regularized sparse auto-encoders

引用

NEUROCOMPUTING 2016年 187卷 59-65页

作者： Liu, Weifeng Ma, Tengzhou Tao, Dapeng You, Jane China Univ Petr Coll Informat & Control Engn Qingdao Peoples R China Yunnan Univ Sch Informat Sci & Engn Kunming Peoples R China Hong Kong Polytech Univ Dept Comp Hong Kong Hong Kong Peoples R China

auto-encoders are one kinds of promising non-probabilistic representation learning paradigms that can efficiently learn stable deterministic features. Recently, auto-encoder algorithms are drawing more and more attentions because of its attractive performance in learning insensitive representation with respect to data changes. The most representative auto-encoder algorithms are the regularized auto-encoders including contractive auto-encoder, denoising auto-encoders, and sparse auto-encoders. In this paper, we incorporate both Hessian regularization and sparsity constraints into auto-encoders and then propose a new auto-encoder algorithm called Hessian regularized sparse auto-encoders (HSAE). The advantages of the proposed HSAE lie in two folds: (1) it employs Hessian regularization to well preserve local geometry for data points;(2) it also efficiently extracts the hidden structure in the data by using sparsity constraints. Finally, we stack the single-layer auto-encoders and form a deep architecture of HSAE. To evaluate the effectiveness, we construct extensive experiments on the popular datasets including MNIST and CIFAR-10 dataset and compare the proposed HSAE with the basic auto-encoders, sparse auto encoders, Laplacian auto-encoders and Hessian auto-encoders. The experimental results demonstrate that HSAE outperforms the related baseline algorithms. (C) 2015 Elsevier B.V. All rights reserved.

关键词： Hessian regularization Sparse representation auto-encoder Manifold

来源：评论

学校读者我要写书评

暂无评论

DeepShape: Deep-Learned Shape Descriptor for 3D Shape Retrieval

引用

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 2017年第7期39卷 1335-1345页

作者： Xie, Jin Dai, Guoxian Zhu, Fan Wong, Edward K. Fang, Yi New York Univ Abu Dhabi Dept Elect & Comp Engn Abu Dhabi 129188 U Arab Emirates NYU Tandon Sch Engn Dept Comp Sci & Engn New York NY 10012 USA

Complex geometric variations of 3D models usually pose great challenges in 3D shape matching and retrieval. In this paper, we propose a novel 3D shape feature learning method to extract high-level shape features that are insensitive to geometric deformations of shapes. Our method uses a discriminative deep auto-encoder to learn deformation-invariant shape features. First, a multiscale shape distribution is computed and used as input to the auto-encoder. We then impose the Fisher discrimination criterion on the neurons in the hidden layer to develop a deep discriminative auto-encoder. Finally, the outputs from the hidden layers of the discriminative auto-encoders at different scales are concatenated to form the shape descriptor. The proposed method is evaluated on four benchmark datasets that contain 3D models with large geometric variations: McGill, SHREC' 10 ShapeGoogle, SHREC' 14 Human and SHREC' 14 Large Scale Comprehensive Retrieval Track Benchmark datasets. Experimental results on the benchmark datasets demonstrate the effectiveness of the proposed method for 3D shape retrieval.

关键词： 3D shape retrieval auto-encoder Feature extraction Fisher discrimination criterion heat diffusion heat kernel signature Heating Kernel Neurons Shape Solid modeling Three-dimensional displays

来源：评论

学校读者我要写书评

暂无评论

Using speech technology for quantifying behavioral characteristics in peer-led team learning sessions

引用

COMPUTER SPEECH AND LANGUAGE 2017年 46卷 343-366页

作者： Dubey, Harishchandra Sangwan, Abhijeet Hansen, John H. L. Univ Texas Dallas Erik Jonsson Sch Engn Comp Sci Ctr Robust Speech Syst Richardson TX 75080 USA

Peer-Led Team Learning (PLTL) is a learning methodology where a peer-leader co-ordinate a small-group of students to collaboratively solve technical problems. PLTL have been adopted for various science, engineering, technology and maths courses in several US universities. This paper proposed and evaluated a speech system for behavioral analysis of PLTL groups. It could help in identifying the best practices for PLTL. The CRSS-PLTL corpus was used for evaluation of developed algorithms. In this paper, we developed a robust speech activity detection (SAD) by fusing the outputs of a DNN-based pitch extractor and an unsupervised SAD based on voicing measures. Robust speaker diarization system consisted of bottleneck features (from stacked autoencoder) and informed HMM-based joint segmentation and clustering system. Behavioral characteristics such as participation, dominance, emphasis, curiosity and engagement were extracted by acoustic analyses of speech segments belonging to all students. We proposed a novel method for detecting question inflection and performed equal error rate analysis on PLTL corpus. In addition, a robust approach for detecting emphasized speech regions was also proposed. Further, we performed exploratory data analysis for understanding the distortion present in CRSS-PLTL corpus as it was collected in naturalistic scenario. The ground-truth Likert scale ratings were used for capturing the team dynamics in terms of student's responses to a variety of evaluation questions. Results suggested the applicability of proposed system for behavioral analysis of small-group conversations such as PLTL, work-place meetings etc.. (C) 2017 Elsevier Ltd. All rights reserved.

关键词： Behavioral speech processing Bottleneck features Curiosity Deep neural network Dominance auto-encoder Emphasis Engagement Peer-led team learning Speaker diarization Small-group conversations

来源：评论

学校读者我要写书评

暂无评论

Supervised source enhancement composed of nonnegative auto-encoders and complementarity subtraction

Supervised source enhancement composed of nonnegative auto-e...

引用

IEEE International Conference on Acoustics, Speech and Signal Processing

作者： Kenta Niwa Yuma Koizumi Tomoko Kawase Kazunori Kobayashi Yusuke Hioka NTT Media Intelligence Laboratories NTT Corporation Japan Department of Mechanical Engineering University of Auckland New Zealand

ISBN: (纸本)9781509041183

A method for constructing deep neural networks (DNNs) for accurate supervised source enhancement is proposed. Attempts were made in previous studies to estimate the power spectral densities (PSDs) of sound sources, which are used to estimate Wiener filters for source enhancement, from the output of multiple beamformings using DNNs. Although performance improved, it was not possible to guarantee accurate PSD estimation since the trained DNNs were treated as black boxes. The proposed DNN construction method uses non-negative auto-encoders and complementarity subtraction. This study also reveals that auto-encoders whose weights are non-negative correspond to non-negative matrix factorization (NMF), which decomposes source PSDs into non-negative spectral bases and their activations. It further introduces a complementarity subtraction method for estimating PSDs accurately. Through several experiments, it was confirmed that the signal-to-interference plus noise ratio improved by approximately 12 dB for datasets captured in various noisy/reverberant rooms.

关键词： microphone array beamforming Wiener filtering deep learning auto-encoder non-negative matrix factorization

来源：评论

学校读者我要写书评

暂无评论

Low-Dose CT With a Residual encoder-Decoder Convolutional Neural Network

引用

IEEE TRANSACTIONS ON MEDICAL IMAGING 2017年第12期36卷 2524-2535页

作者： Chen, Hu Zhang, Yi Kalra, Mannudeep K. Lin, Feng Chen, Yang Liao, Peixi Zhou, Jiliu Wang, Ge Sichuan Univ Coll Comp Sci Chengdu 610065 Sichuan Peoples R China Massachusetts Gen Hosp Dept Radiol Boston MA 02114 USA Southeast Univ Lab Image Sci & Technol Nanjing 210096 Jiangsu Peoples R China Southeast Univ Key Lab Comp Network & Informat Integrat Minist Educ Nanjing 210096 Jiangsu Peoples R China Sixth Peoples Hosp Chengdu Dept Sci Res & Educ Chengdu 610065 Sichuan Peoples R China Rensselaer Polytech Inst Dept Biomed Engn Troy NY 12180 USA

Given the potential risk of X-ray radiation to the patient, low-dose CT has attracted a considerable interest in the medical imaging field. Currently, the main stream low-dose CT methods include vendor-specific sinogram domain filtration and iterative reconstruction algorithms, but they need to access raw data, whose formats are not transparent to most users. Due to the difficulty of modeling the statistical characteristics in the image domain, the existing methods for directly processing reconstructed images cannot eliminate image noise very well while keeping structural details. Inspired by the idea of deep learning, here we combine the autoencoder, deconvolution network, and shortcut connections into the residual encoder-decoder convolutional neural network (RED-CNN) for low-dose CT imaging. After patch-based training, the proposed RED-CNN achieves a competitive performance relative to the-state-of-art methods in both simulated and clinical cases. Especially, our method has been favorably evaluated in terms of noise suppression, structural preservation, and lesion detection.

关键词： Low-dose CT deep learning auto-encoder convolutional deconvolutional residual neural network

来源：评论

学校读者我要写书评

暂无评论

An improved vector quantization method using deep neural network

引用

AEU-INTERNATIONAL JOURNAL OF ELECTRONICS AND COMMUNICATIONS 2017年 72卷 178-183页

作者： Jiang, Wenbin Liu, Peilin Wen, Fei Shanghai Jiao Tong Univ Dept Elect Engn Shanghai 200240 Peoples R China Air Force Engn Univ Air Control & Nav Inst Xian 710000 Peoples R China

To address the challenging problem of vector quantization (VQ) for high dimensional vector using large coding bits, this work proposes a novel deep neural network (DNN) based VQ method. This method uses a k-means based vector quantizer as an encoder and a DNN as a decoder. The decoder is initialized by the decoder network of deep auto-encoder, fed with the codes provided by the k-means based vector quantizer, and trained to minimize the coding error of VQ system. Experiments on speech spectrogram coding demonstrate that, compared with the k-means based method and a recently introduced DNN-based method, the proposed method significantly reduces the coding error. Furthermore, in the experiments of coding multi-frame speech spectrogram, the proposed method achieves about 11% relative gain over the k-means based method in terms of segmental signal to noise ratio (SegSNR). (C) 2016 Elsevier GmbH. All rights reserved.

关键词： Deep neural network Vector quantization auto-encoder Binary coding

来源：评论

学校读者我要写书评

暂无评论

automatic feature extraction of time-series applied to fault severity assessment of helical gearbox in stationary and non-stationary speed operation

引用

APPLIED SOFT COMPUTING 2017年 58卷 53-64页

作者： Cabrera, Diego Sancho, Fernando Li, Chuan Cerrada, Mariela Sanchez, Rene-Vinicio Pacheco, Fannia de Oliveira, Jose Valente Univ Politecn Salesiana Sede Cuenca Dept Mech Engn Cuenca Ecuador Univ Seville Dept Comp Sci & Artificial Intelligence Seville Spain Chongqing Technol & Business Univ Chongqing Key Lab Mfg Equipment Mech Design & Con Chongqing Peoples R China Univ Los Andes CEMISID Merida Venezuela Univ Algarve CEOT Faro Portugal

Signals captured in rotating machines to obtain the status of their components can be considered as a source of massive information. In current methods based on artificial intelligence to fault severity assessment, features are first generated by advanced signal processing techniques. Then feature selection takes place, often requiring human expertise. This approach, besides time-consuming, is highly dependent on the machinery configuration as in general the results obtained for a mechanical system cannot be reused by other systems. Moreover, the information about time events is often lost along the process, preventing the discovery of faulty state patterns in machines operating under time-varying conditions. In this paper a novel method for automatic feature extraction and estimation of fault severity is proposed to overcome the drawbacks of classical techniques. The proposed method employs a Deep Convolutional Neural Network pre-trained by a Stacked Convolutional autoencoder. The robustness and accuracy of this new method are validated using a dataset with different severity conditions on failure mode in a helical gearbox, working in both constant and variable speed of operation. The results show that the proposed unsupervised feature extraction method is effective for the estimation of fault severity in helical gearbox, and it has a consistently better performance in comparison with other reported feature extraction methods. (C) 2017 Elsevier B.V. All rights reserved.

关键词： Deep learning Convolution auto-encoder Wavelet packets Helical gearbox

来源：评论

学校读者我要写书评

暂无评论

Cluster Naturalistic Driving Encounters Using Deep Unsupervised Learning

Cluster Naturalistic Driving Encounters Using Deep Unsupervi...

引用

IEEE Intelligent Vehicles Symposium

作者： Sisi Li Wenshuo Wang Zhaobin Mo Ding Zhao Robotics Institute University of Michigan Ann Arbor MI Department of Mechanical Engineering University of Michigan Ann Arbor Automotive Engineering at the Tsinghua University Beijing China

Learning knowledge from driving encounters could help self-driving cars make appropriate decisions when driving in complex settings with nearby vehicles engaged. This paper develops an unsupervised classifier to group naturalistic driving encounters into distinguishable clusters by combining an auto-encoder with k-means clustering (AE-kMC). The effectiveness of AE-kMC was validated using the data of 10,000 naturalistic driving encounters which were collected by the University of Michigan, Ann Arbor in the past five years. We compare our developed method with the k-means clustering methods and experimental results demonstrate that the AE-kMC method outperforms the original k-means clustering method.

关键词： Driving encounter classification Unsupervised learning auto-encoder

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：