检索结果-内蒙古大学图书馆

2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops

作者： Anan, Liu Sheng, Tang Yongdong, Zhang Yan, Song Jintao, Li Zhaoxuan, Yang Virtual Reality Laboratory Institute of Computing Technology Chinese Academy of Sciences Beijing 100190 China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing 100190 China Department of Electronic Engineering Tianjin University Tianjin 300072 China

ISBN: (纸本)9781424423408

In this paper, we specially propose a hierarchical framework for movie content analysis. The purpose of our work is trying to realize computers' understanding for movie content, especially "Who, What, Where, How" which occur in the storyline by imitating human perception and cognition. The framework consists of two hierarchies. As for the low level part, we originally construct the human attention model with temporal information motivated by the Weber-Fechner Law to depict the variation of human perception in multiple modalities. As for the high level part, we focus on semantic understanding of different granularities of videos and simulate human cognition for movie content. Based on this hierarchical framework, we present its applications on semantic retrieval, video summarization and content filter. The promising results of users' subjective assessment indicate that the proposed framework is applicable for automatic analysis of movie content by computers. © 2008 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Symmetrical Invariant LBP Texture Descriptor and Application for Image Retrieval

Symmetrical Invariant LBP Texture Descriptor and Application...

引用

Congress on Image and Signal processing, CISP

作者： Zhiping Shi Fei Ye Qing He Zhongzhi Shi Key Laboratory of Intelligent Information Processing Institute of Computing Technology CAS Beijing China

In this paper, we present a symmetrical invariant LBP (SILBP) texture descriptor based on the texture pattern equivalent, which is defined according to visual perception of textures. Then we extend the SILBP to linear SILBP. Our experiments on image retrieval show that the proposed texture discriptor has the advantages of symmetrical invariant, rotation robustness and computing simplicity.

关键词： Image retrieval Histograms Visual perception Robustness Humans Image texture Content based retrieval Wavelet transforms Signal processing information retrieval

来源：评论

学校读者我要写书评

暂无评论

The Data Selection Criteria for HSC and SVM Algorithms

The Data Selection Criteria for HSC and SVM Algorithms

引用

International Conference on Natural Computation (ICNC)

作者： Qing He Fuzhen Zhuang Zhongzhi Shi The Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences China

This paper makes a discussion of consistent subsets (CS) selection criteria for hyper surface Classification (HSC) and SVM algorithms. The consistent subsets play an important role in the data selection. Firstly, the paper proposes that minimal consistent subset for a disjoint cover set (MCSC) plays an important role in the data selection for HSC. The MCSC can be applied to select a representative subset from the original sample set for HSC. MCSC has the same classification model with the entire sample set and can totally reflect its classification ability. Secondly, the number of MCSC is calculated. Thirdly, by comparing the performance of HSC and SVM on corresponding CS, we argue that it is not reasonable that using the same train data set to train different classifiers and then testing the classifiers by the same test data set for different algorithms. The experiments show that algorithms can respectively select the proper data set for training, which ensures good performance and generalization ability. MCSC is the best selection for HSC, and support vector set is the effective selection for SVM.

关键词： Support vector machines Support vector machine classification Nearest neighbor searches Neural networks Testing Laboratories information processing Computers Helium Databases

来源：评论

学校读者我要写书评

暂无评论

Personalized event-based news video retrieval with dynamic user-log

Personalized event-based news video retrieval with dynamic u...

引用

IEEE International Conference on Multimedia and Expo (ICME)

作者： Ming Li Yantao Zheng Shi-Yong Neo Xiangdong Wang Sheng Tang Shou-Xun Lin School of Computing NUS Singapore Key Laboratory of Intelligent Information Processing ICT CAS China

Personalization especially in the domain of information retrieval is essentially important, as users might pose the same query even when they are searching for different information. It is thus necessary to create a retrieval engine which takes into consideration the dynamic information needs of different users. This paper presents our personalized news video retrieval engine, which exploits the individual userpsilas previous browsing history to customize and enhance their future search results. Specifically, the system utilizes the news topic hierarchy, a hierarchical news topic structure derived from unsupervised clustering on the news video corpus and event entities from news video and online news articles. We then dynamically project userpsilas browsing history onto this topic hierarchy to provide the basis for re-ranking relevant news videos. This system is tested on one month of TRECVID 2006 dataset consisting of 80 hours news video and found to return results in a more intuitive and personalized manner.

关键词： History Speech Visualization Search engines Speech recognition Explosions Streaming media

来源：评论

学校读者我要写书评

暂无评论

Transfer learning from multiple source domains via consensus regularization

Transfer learning from multiple source domains via consensus...

引用

17th ACM Conference on information and Knowledge Management, CIKM'08

作者： Luo, Ping Zhuang, Fuzhen Xiong, Hui Xiong, Yuhong He, Qing HP Labs China China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences China Graduate University of Chinese Academy of Sciences China MSIS Department Rutgers University United States

ISBN: (纸本)9781595939913

Recent years have witnessed an increased interest in transfer learning. Despite the vast amount of research performed in this field, there are remaining challenges in applying the knowledge learnt from multiple source domains to a target domain. First, data from multiple source domains can be semantically related, but have different distributions. It is not clear how to exploit the distribution differences among multiple source domains to boost the learning performance in a target domain. Second, many real-world applications demand this transfer learning to be performed in a distributed manner. To meet these challenges, we propose a consensus regularization framework for transfer learning from multiple source domains to a target domain. In this framework, a local classifier is trained by considering both local data available in a source domain and the prediction consensus with the classifiers from other source domains. In addition, the training algorithm can be implemented in a distributed manner, in which all the source-domains are treated as slave nodes and the target domain is used as the master node. To combine the training results from multiple source domains, it only needs share some statistical data rather than the full contents of their labeled data. This can modestly relieve the privacy concerns and avoid the need to upload all data to a central location. Finally, our experimental results show the effectiveness of our consensus regularization learning. Copyright 2008 ACM.

关键词： Classification Consensus regularization Transfer learning

来源：评论

学校读者我要写书评

暂无评论

COUNT PASSENGERS BASED ON HAAR-LIKE FEATURE IN ELEVATOR APPLICATION

COUNT PASSENGERS BASED ON HAAR-LIKE FEATURE IN ELEVATOR APPL...

引用

2008 International Conference on Machine Learning and Cybernetics(2008机器学习与控制论国际会议)

作者： HONG LIU YUE-LIANG QIAN QUN LIU JIN-TAO LI Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing 100080

Present elevator control use button sensors to determine when and where to dispatch an elevator car, which don't use the number of passengers. In this paper, we analyze images from camera to detect how many persons waiting for the elevator or in an elevator. A novel framework is proposed for optimized elevator schedule. Extended Haar-like features and Adaboost are used to train a head-shoulder classifier. Some images are selected from video according to elevator button callings to detect head-shoulder. To reduce false alarms a post process is added after detecting. Experimental results show the proposed method with post process has higher performance than existed methods. The information of passenger number can be send to elevator control system for effective schedule, which can reduce passengers waiting time and elevator's unnecessary stop, finally save energy and reduce maintain fee.

关键词： Elevator control system Haar-like features Adaboost Post process

来源：评论

学校读者我要写书评

暂无评论

An Adaptive Travel Itinerary Selection Assistant under Fuzzy User Preferences

An Adaptive Travel Itinerary Selection Assistant under Fuzzy...

引用

2008 Conference on Human System Interactions (HSI 2008), vol.1

作者： Ruihong Shi Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China

This paper presents a candidate-evaluation model (CEM) which interactively elicits user preferences and assists decision makers in decision making in applications such as travel itinerary planning. The CEM contrasts with traditional decision analytic and planning frameworks in which a complete user model is elicited beforehand or is constructed by a human expert. We used the CEM model to implement an Itinerary Selection Assistant (ISA) system, which helps tourists identify satisfactory travel itineraries. The ISA starts with fuzzy user preferences and gradually approximate the optimal solution through carefully choosing candidate solutions to present to the user and inferring user's actual preferences by analyzing user evaluations over the candidates.

关键词： Augmented weighted Tchebycheff programming Fuzzy user preferences Multi-attribute decision making Travel itinerary User modeling

来源：评论

学校读者我要写书评

暂无评论

The Maximum Entropy based Rule Selection Model for Statistical Machine Translation (Invited Paper)

The Maximum Entropy based Rule Selection Model for Statistic...

引用

International Symposium on Universal Communication, ISUC

作者： Qun Liu Zhongjun He Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China

This paper presents a novel rule selection model for statistical machine translation (SMT) that uses the maximum entropy approach to predict target-side for an ambiguous source-side. The maximum entropy based rule selection (MERS) model combines rich contextual information as features, thus can help SMT systems perform context-dependent rule selection. We incorporate the MERS model into two kinds of the state-of-the-art syntax-based SMT models: the hierarchical phrase-based model and the tree-to-string alignment template model. Experiments show that our approach achieves significant improvements over both the baseline systems.

关键词： Entropy Surface-mount technology Helium Decoding Predictive models Context modeling Power generation economics Laboratories Machine intelligence information processing

来源：评论

学校读者我要写书评

暂无评论

Focused Crawling with Heterogeneous Semantic information

Focused Crawling with Heterogeneous Semantic Information

引用

IEEE WIC ACM International Conference on Web Intelligence (WI)

作者： Rui Huang Fen Lin Zhongzhi Shi Chinese Academy of Sciences Beijing China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy and Sciences Beijing China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy and Sciences Beijing China

Focused crawlers selectively retrieve Web documents that are relevant to a predefined set of topics. To intelligently make predictions and decisions about relevant URLs and web pages, different topic models have been introduced to represent topic-specific knowledge. Yet it is difficult to support semantic interoperability among different models. Moreover, some manually specified additional semantic information, such as semantic markups and social annotations, could not be effectively used to improve crawling. This paper proposes to boost focused crawling with four kinds of semantic models and semantic information, including thesauruses, categories, ontologies, and folksonomies. A statistical semantic association model is proposed to integrate different semantic models, represent heterogeneous semantic information, and support semantic relevance computation. A focused crawling framework is developed which adopts both keyword based contents and different kinds of additional information for relevance prediction and ranking. Experiments show that the proposed model and framework effectively integrates heterogeneous semantic information for focused crawling.

关键词： Crawlers Web pages Uniform resource locators intelligent agent Computers Ontologies information processing Semantic Web information retrieval Predictive models

来源：评论

学校读者我要写书评

暂无评论

New approach to heterogeneous semantic search on the web

引用

Jisuanji Yanjiu yu Fazhan/Computer Research and Development 2008年第8期45卷 1338-1345页

作者： Huang, Rui Shi, Zhongzhi Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing 100190 China Graduate University Chinese Academy of Sciences Beijing 100049 China

Relevance ranking is a key to Web search in determining how results are retrieved and ordered. As keyword-based search does not guarantee relevance in meanings, semantic search has been put forward as an attractive and promising approach. Recently several kinds of semantic information have been adopted in search respectively, such as thesauruses, ontologies and semantic markups, as well as folksonomies and social annotations. However, although to integrate more semantics would logically generate better search results, search mechanism to fully adopt different kinds of semantic information is still in absence and to be researched. To these ends, an integrated semantic search mechanism is proposed to incorporate textual information and keyword search with heterogeneous semantic information and semantic search. A statistical based measurement of semantic relevance, defined as semantic probabilities, is introduced to integrate both keywords and four kinds of semantic information including thesauruses, categories, ontologies and folksonomies. It is calculated with all textual information and semantic information, and stored in a newly proposed index structure called semantic-keyword dual index. Based on this uniform measurement, the search mechanism is developed that fully utilizes existing keyword and semantic search mechanisms to enhance heterogeneous semantic search. Experiments show that the proposed approach can effectively integrate both keyword-based information and heterogeneous semantic information in search.

关键词： Semantic Web

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：