检索结果-内蒙古大学图书馆

4th International Conference on Internet multimedia computing and Service, ICIMCS 2012

作者： Qin, Lei Cheng, Zhongwei Huang, Qingming Pang, Junbiao Key Lab. of Intel. Inf. Proc. Institute of Computing Technology Chinese Academy of Sciences Beijing China Graduate University of Chinese Academy of Sciences Beijing China Beijing Municipal Key Lab. of Multimedia and Intelligent Software Technology Beijing Univ. of Tech. China

ISBN: (纸本)9781450316002

As an important aspect in video content analysis, event detection is still an open problem. In particular, the study on detecting interactive events in crowd scenes is still limited. In this paper, we investigate detecting interactive events between persons, e.g. PeopleMeet, PeopleSplitUp and Embrace in complex scenes using a sequence learning based approach. By sequence learning, the spatial-temporal context information is introduced in the learning stage. Experiments have been performed over TRECVid Event Detection 2010 dataset, which contains totally 144 hours surveillance video of London Gatwick airport. According to the TRECVid-ED 2010 formal evaluation, our approach obtains promising results, with the top performance (NDCR) for PeopleMeet and PeopleSplit-Up, and second-best performance (NDCR) for Embrace. Copyright © 2012 ACM.

关键词： Security systems

来源：评论

学校读者我要写书评

暂无评论

Automatic video genre categorization and event detection techniques on large-scale sports data 10

Automatic video genre categorization and event detection tec...

引用

Proceedings of the 2010 Conference of the Center for Advanced Studies on Collaborative Research

作者： Zhang, Ning Duan, Ling-Yu Huang, Qingming Li, Lingfang Gao, Wen Guan, Ling Ryerson Multimedia Research Laboratory Ryerson University Toronto ON Canada Institute of Digital Media Peking University Beijing 100871 China Key Lab. of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing 100080 China Graduate University of Chinese Academy of Sciences Beijing 100190 China

This paper presents an efficient and robust automatic process for large-scale sports video analysis. The proposed system firstly identifies the genre of the query video, and then accomplishes the interesting event detection task. The significance of this framework is its automatic characteristic in testing with minimum human involvement in training, as well as the scalability and expansibility in dealing with a large-scale dataset. Domain-knowledge independent local features are extracted from an input video sequence and a histogram based distribution representation is created using the bag-of-visual-words (BoW) model. In genre categorization, k-nearest neighbor (k-NN) classifiers with various dissimilarity measures are assessed and evaluated analytically. For the event detection, a hidden conditional random field (HCRF) structured prediction model is utilized. Overall, this framework demonstrates the efficiency and accuracy in processing voluminous data from sports collection and achieves various tasks in video analysis. It also demonstrates a potential technology transformation from the "laboratory bench" to commercial applications. © 2010 Ning Zhang, Ling-Yu Duan, Qing-ming Huang, Lingfang Li, Wen Gao and Ling Guan.

关键词： Sports

来源：评论

学校读者我要写书评

暂无评论

Combining Deep Embeddings of Acoustic and Articulatory Features for Speaker Identification

Combining Deep Embeddings of Acoustic and Articulatory Featu...

引用

IEEE International Conference on Acoustics, Speech and Signal Processing

作者： Qian-Bei Hong Chung-Hsien Wu Hsin-Min Wang Chien-Lin Huang Graduate Program of Multimedia Systems and Intelligent Computing National Cheng Kung University and Academia Sinica Tainan Taiwan PingAn AI Lab Palo Alto CA USA

ISBN: (数字)9781509066315

ISBN: (纸本)9781509066322

In this study, deep embedding of acoustic and articulatory features are combined for speaker identification. First, a convolutional neural network (CNN)-based universal background model (UBM) is constructed to generate acoustic feature (AC) embedding. In addition, as the articulatory features (AFs) represent some important phonological properties during speech production, a multilayer perceptron (MLP)-based AF embedding extraction model is also constructed for AF embedding extraction. The extracted AC and AF embeddings are concatenated as a combined feature vector for speaker identification using a fully-connected neural network. This proposed system was evaluated by three corpora consisting of King-ASR, LibriSpeech and SITW, and the experiments were conducted according to the properties of the datasets. We adopted all three corpora to evaluate the effect of AF embedding, and the results showed that combining AF embedding into the input feature vector improved the performance of speaker identification. The LibriSpeech corpus was used to evaluate the effect of the number of enrolled speakers. The proposed system achieved an EER of 7.80% outperforming the method based on x-vector with PLDA (8.25%). And we further evaluated the effect of signal mismatch using the SITW corpus. The proposed system achieved an EER of 25.19%, which outperformed the other baseline methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Statistics Pooling Time Delay Neural Network Based on X-Vector for Speaker Verification

Statistics Pooling Time Delay Neural Network Based on X-Vect...

引用

IEEE International Conference on Acoustics, Speech and Signal Processing

ISBN: (数字)9781509066315

ISBN: (纸本)9781509066322

This paper aims to improve speaker embedding representation based on x-vector for extracting more detailed information for speaker verification. We propose a statistics pooling time delay neural network (TDNN), in which the TDNN structure integrates statistics pooling for each layer, to consider the variation of temporal context in frame-level transformation. The proposed feature vector, named as statsvector, are compared with the baseline x-vector features on the VoxCeleb dataset and the Speakers in the Wild (SITW) dataset for speaker verification. The experimental results showed that the proposed stats-vector with score fusion achieved the best performance on VoxCeleb1 dataset. Furthermore, considering the interference from other speakers in the recordings, we found that the proposed statsvector efficiently reduced the interference and improved the speaker verification performance on the SITW dataset.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Advances in multimedia Information Processing – PCM 2014 1

引用

丛书名： Lecture Notes in Computer Science

1000年

作者： Wei Tsang Ooi Cees G. M. Snoek Hung Khoon Tan Chin-Kuan Ho Benoit Huet Chong-Wah Ngo

ISBN: (数字)9783319131689

ISBN: (纸本)9783319131672

This book constitutes the refereed proceedings of the 15th Pacific Rim Conference on multimedia, PCM 2014, held in Kuching, Malaysia, in December 2014. The 35 revised full papers and 6 short papers presented were carefully reviewed and selected from 84 submissions. The papers cover a wide range of topics in the area of multimedia content analysis, multimedia signal processing and communications, and multimedia applications and services. They have been organized into topical sections on video coding, annotation, image and photo, applications, people, image analysis and processing under extra help, nearest neighbor, neural networks, and audio. Also included are sections with best papers and posters and demonstrations.

关键词： multimedia Information Systems Information Systems Applications (incl. Internet) Image Processing and Computer Vision Pattern Recognition Data Mining and Knowledge Discovery User Interfaces and Human Computer Interaction

来源：评论

学校读者我要写书评

暂无评论

MFPR: A Personalized Ranking Recommendation with Multiple Feedback

引用

ACM Transactions on Social computing 2018年第2期1卷 1–22页

作者： Shi, Chuan Liu, Jian Zhang, Yiding Hu, Binbin Liu, Shenghua Yu, Philip S. Beijing Key Lab of Intelligent Telecommunication Software and Multimedia Beijing University of Posts and Telecommunications No. 10 Xi Tu Cheng Road Beijing China Cas Key Laboratory of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences No.6 Kexueyuan South Road Zhongguancun Beijing China Department of Computer Science University of Illinois at Chicago 851 South Morgan Street Chicago United States

Recently, recommender systems have played an important role in improving web user experiences and increasing profits. Recommender systems exploit users' behavioral history (i.e., feedback on items) to build models. The feedback usually includes explicit feedback (e.g., ratings) and implicit feedback (e.g., browsing history, click logs), which are both useful for improving recommendations. However, as far as we are concerned, no existing works have integrated both explicit and multiple implicit feedback simultaneously. Therefore, we propose a unified and flexible model, named Multiple Feedback-based Personalized Ranking (MFPR), to make full use of multiple feedback, which uses a personalized ranking framework. To train model MFPR, we design an algorithm to generate ordered item pairs as labeled data, with consideration of both rating scores and multiple implicit feedback. Extensive experiments on two real-world datasets validate the effectiveness of the MFPR model. With the integration of multiple feedback, MFPR significantly improves recommendation performance. © 2018 ACM.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Joint Multiple Dictionary Learning for Tensor Sparse Coding

Joint Multiple Dictionary Learning for Tensor Sparse Coding

引用

International Joint Conference on Neural Networks

作者： Yifan Fu Junbin Gao Yanfeng Sun Xia Hong School of Computing and Mathematics Charles Sturt University Beijing Municipal Key Lab of Multimedia and Intelligent Software Technology Beijing University of Technology School of Systems Engineering University of Reading

ISBN: (纸本)9781479914821

Traditional dictionary learning algorithms are used for finding a sparse representation on high dimensional data by transforming samples into a one-dimensional (1D) vector. This 1D model loses the inherent spatial structure property of data. An alternative solution is to employ Tensor Decomposition for dictionary learning on their original structural form - a tensor - by learning multiple dictionaries along each mode and the corresponding sparse representation in respect to the Kronecker product of these dictionaries. To learn tensor dictionaries along each mode, all the existing methods update each dictionary iteratively in an alternating manner. Because atoms from each mode dictionary jointly make contributions to the sparsity of tensor, existing works ignore atoms correlations between different mode dictionaries by treating each mode dictionary independently. In this paper, we propose a joint multiple dictionary learning method for tensor sparse coding, which explores atom correlations for sparse representation and updates multiple atoms from each mode dictionary simultaneously. In this algorithm, the Frequent-Pattern Tree (FP-tree) mining algorithm is employed to exploit frequent atom patterns in the sparse representation. Inspired by the idea of K-SVD, we develop a new dictionary update method that jointly updates elements in each pattern. Experimental results demonstrate our method outperforms other tensor based dictionary learning algorithms.

关键词： Tensile stress Dictionaries Joints Vectors Correlation Databases Encoding

来源：评论

学校读者我要写书评

暂无评论

Active Learning for Efficient Sampling of Control Models of Collectives

Active Learning for Efficient Sampling of Control Models of ...

引用

International Conference on Self-Adaptive and Self-Organizing Systems (SASO)

作者： Alexander Schiendorfer Christoph Lassner Gerrit Anders Wolfgang Reif Rainer Lienhart Institute for Software & Systems Engineering University of Augsburg Germany Perceiving Systems Department Max Planck Institute for Intelligent Systems Germany Multimedia Computing and Computer Vision Lab University of Augsburg Germany

Many large-scale systems benefit from an organizational structure to provide for problem decomposition. A pivotal problem solving setting is given by hierarchical control systems familiar from hierarchical task networks. If these structures can be modified autonomously by, e.g., Coalition formation and reconfiguration, adequate decisions on higher levels require a faithful abstracted model of a collective of agents. An illustrative example is found in calculating schedules for a set of power plants organized in a hierarchy of Autonomous Virtual Power Plants. Functional dependencies over the combinatorial domain, such as the joint costs or rates of change of power production, are approximated by repeatedly sampling input-output pairs and substituting the actual functions by piecewise linear functions. However, if the sampled data points are weakly informative, the resulting abstracted high-level optimization introduces severe errors. Furthermore, obtaining additional point labels amounts to solving computationally hard optimization problems. Building on prior work, we propose to apply techniques from active learning to maximize the information gained by each additional point. Our results show that significantly better allocations in terms of cost-efficiency (up to 33.7 % reduction in costs in our case study) can be found with fewer but carefully selected sampling points using Decision Forests.

关键词： Power generation Production Cost function Schedules Robots Switches

来源：评论

学校读者我要写书评

暂无评论

Automatic sports genre categorization and view-type classification over large-scale dataset 09

Automatic sports genre categorization and view-type classifi...

引用

17th ACM International Conference on multimedia, MM'09, with Co-located Workshops and Symposiums

作者： Li, Lingfang Zhang, Ning Duan, Ling-Yu Huang, Qingming Du, Jun Guan, Ling Key Lab. of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing 100080 China Ryerson Multimedia Research Laboratory Ryerson University Toronto ON Canada Institute of Digital Media Peking University Beijing 100871 China Graduate University of Chinese Academy of Sciences Beijing 100190 China NEC Research Labs. China Beijing 100084 China

ISBN: (纸本)9781605586083

This paper presents a framework with two automatic tasks targeting large-scale and low quality sports video archives collected from online video streams. The framework is based on the bag of visual-words model using speeded-up robust features (SURF). The first task is sports genre categorization based on hierarchical structure. Following on the second task which is based on automatically obtained genre, views are classified using support vector machines (SVMs). As a consequence, the views classification result can be used in video parsing and highlight extraction. As compared with state-of-the-art methods, our approach is fully automatic as well as domain knowledge free and thus provides a better extensibility. Furthermore, our dataset consists of 14 sport genres with 6850 minutes in total. Both sport genre categorization and view type classification have more than 80% accuracy rates, which validate this framework's robustness and potential in web-based applications. Copyright 2009 ACM.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Emotion Regulation in intelligent Tutoring Systems: A Systematic Literature Review

Emotion Regulation in Intelligent Tutoring Systems: A System...

引用

IEEE International Conference on Teaching, Assessment and Learning for Engineering (TALE)

作者： Rihab Khadimallah Mouna Abdelkefi Ilhem Kallel REGIM-Lab: Research Groups in Intelligent Machines ISIMS: Higher Institute of Computer Science and Multimedia University of Sfax Tunisia MIRACL: Multimedia InfoRmation Systems and Advanced Computing Laboratory ENIS: National School of Engineers Route Soukra Km 3.5 BP 1173 3038 University of Sfax Tunisia

ISBN: (数字)9781728169422

ISBN: (纸本)9781728169439

Emotion regulation (ER) is a process responsible for altering emotions' reactions and expressions as well as situations that simulate such feelings to better respond to environmental ***, ER is actually a mere mental and behavioral process that many of us already perform consciously and unconsciously. In this issue, with the imperious role of emotions in learning process which they crucially influence student's cognitive performances and academic outcomes, several researchers proposed different methods and tools for regulating learners' negative emotions while learning in online environment. In the present paper, we discuss the current and original research works that apply different strategies and methods for emotion regulation in different smart environments (i.e., intelligent Tutoring Systems) by adopting a Systematic Literature Review (SLR) approach.

关键词： Systematics Bibliographies Education Machine learning Tools Regulation Erbium

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：