检索结果-内蒙古大学图书馆

International Symposium on Distributed Computing and Applications to Business, Engineering and Science (DCABES)

作者： Jian Wu Heng-jun Yue Yan-yan Cao Zhi-ming Cui Provincial Key Laboratory of Computer Information Processing Technology Soochow University Suzhou China Soochow University Suzhou Jiangsu CN

Combing with specific temporal information of video, this paper proposes a kind of video object tracking method based on normalized cross-correlation matching by using the high precision characteristics of normalized cross-correlation image matching. Firstly, extract video background from the temporal information of video. Then, acquire the region of moving object using background subtraction. Lastly, carry out related matching and updating towards the extracted moving object by means of normalized cross-correlation. Experimental result shows that the adaptability of our method is strong, which can well solve the tracking problems when tracking objects have scale transform. It also has good anti-interference ability and robustness, and can track moving objects accurately under the condition of noise interference, lens dithering and background mutation.

关键词： Target tracking Noise Correlation Pixel Vehicles Cameras

来源：评论

学校读者我要写书评

暂无评论

The application of flash into M-learning in 3G initial stage

The application of flash into M-learning in 3G initial stage

引用

International Symposium on Knowledge Acquisition and Modeling, KAM

作者： Yue-ming Tang Quan Liu Ling-ling Wang Department of Educational Technology Henan Normal University Xinxiang Henan China Institute of Computer Science and Technology Provincial Key Laboratory of Computer Information Processing Technology Soochow University Suzhou Jiangsu China

With the development of the 3G technology, M-learning enters a period of rapid development. But meanwhile 3G also restrains the development of M-learning due to several drawbacks in its early days. According to the application analysis, it comes out Flash has its unique features: small capacity, easy to develop high-quality and etc, which can make up for 3G initial stage's disadvantages. Through case design, this paper proves that Flash can be a new effective approach and way for M-learning in the 3G network. In the end, from the aspects of technical and artistic, this paper makes a further research on how to improve the flash animation to serve the M-learning better.

关键词： Multimedia communication Education Animation

来源：评论

学校读者我要写书评

暂无评论

3D FACE REPRESENTATION AND RECOGNITION BY INTRINSIC SHAPE DESCRIPTION MAPS

3D FACE REPRESENTATION AND RECOGNITION BY INTRINSIC SHAPE DE...

引用

IEEE International Conference on Acoustics, Speech, and Signal processing

作者： Zhe Guo Yanning Zhang Yong Xia Zenggang Lin Dagan Feng Shaanxi Provincial Key Lab of Speech & Image Information Processing (SAIIP) School of Computer Science Northwestern Polytechnical University Xi'an China Biomedical and Multimedia Information Technology (BMIT) Research Group School of Information Technologies University of Sydney Australia

ISBN: (纸本)9781424442959

We present a novel method for 3D face recognition, in which the 3D facial surface is first mapped into a 2D domain with specified resolution through a global optimization by constrained conformal geometric maps. The Intrinsic Shape Description Map (ISDM) is then constructed through a modeling technique capable to express geometric and appearance information of the 3D face. Hence the 3D surface matching problem can be simplified to a 2D image matching problem, which greatly reduces the computational complexity. Finally, the Intrinsic Shape Description Feature (ISDF) of ISDM and the discrimination analysis can be calculated. Experimental results implemented on GavabDB demonstrate that our proposed method significantly outperforms the existing methods with respect to pose variation.

关键词： Face recognition Conformal Geometric Maps Intrinsic Shape Description Map

来源：评论

学校读者我要写书评

暂无评论

A peer selection prior data scheduling method for peer-to-peer media streaming system

引用

Beijing Gongye Daxue Xuebao/Journal of Beijing University of technology 2010年第SUPPL. 1期36卷 26-30页

作者： Ji, Qi-Jin Yang, Zhe Zhu, Yan-Qin School of Computer Science and Technology Soochow University Suzhou 215006 Jiangsu China Provincial Key Laboratory for Computer Information Processing Technology Soochow University Suzhou 215006 Jiangsu China

Data Scheduling in P2P media streaming system considers which peer to select for interesting media data. Since data scheduling decision can only be made in terms of the local information of each peer, the commonly used data block selection prior data scheduling is prone to traffic concentration to minor peers in the system and thus degrades the system throughput greatly. A peer selection prior data scheduling method is proposed in this paper, which can direct traffic to the neighbors evenly. The simulation results show that the proposed method can mitigate the load of source server more than 25% and meanwhile improve the system throughput comparing with the data block selection prior alternatives.

关键词： Peer to peer networks

来源：评论

学校读者我要写书评

暂无评论

Global learning of noun phrase anaphoricity in coreference resolution via lab.l propagation

Global learning of noun phrase anaphoricity in coreference r...

引用

2009 Conference on Empirical Methods in Natural Language processing, EMNLP 2009, Held in Conjunction with ACL-IJCNLP 2009

作者： Zhou, Guo Dong Kong, Fang JiangSu Provincial Key Lab. for Computer Information Processing Technology School of Computer Science and Technology Soochow University Suzhou 215006 China

Knowledge of noun phrase anaphoricity might be profitably exploited in coreference resolution to bypass the resolution of non-anaphoric noun phrases. However, it is surprising to notice that recent attempts to incorporate automatically acquired anaphoricity information into coreference resolution have been somewhat disappointing. This paper employs a global learning method in determining the anaphoricity of noun phrases via a lab.l propagation algorithm to improve learning-based coreference resolution. In particular, two kinds of kernels, i.e. the feature-based RBF kernel and the convolution tree kernel, are employed to compute the anaphoricity similarity between two noun phrases. Experiments on the ACE 2003 corpus demonstrate the effectiveness of our method in anaphoricity determination of noun phrases and its application in learning-based coreference resolution. © 2009 ACL and AFNLP.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Employing the centering theory in pronoun resolution from the semantic perspective

Employing the centering theory in pronoun resolution from th...

引用

2009 Conference on Empirical Methods in Natural Language processing, EMNLP 2009, Held in Conjunction with ACL-IJCNLP 2009

作者： Kong, Fang Zhou, Guo Dong Zhu, Qiaoming JiangSu Provincial Key Lab. for Computer Information Processing Technology School of Computer Science and Technology Soochow University Suzhou 215006 China

In this paper, we employ the centering theory in pronoun resolution from the semantic perspective. First, diverse semantic role features with regard to different predicates in a sentence are explored. Moreover, given a pronominal anaphor, its relative ranking among all the pronouns in a sentence, according to relevant semantic role information and its surface position, is incorporated. In particular, the use of both the semantic role features and the relative pronominal ranking feature in pronoun resolution is guided by extending the centering theory from the grammatical level to the semantic level in tracking the local discourse focus. Finally, detailed pronominal subcategory features are incorporated to enhance the discriminative power of both the semantic role features and the relative pronominal ranking feature. Experimental results on the ACE 2003 corpus show that the centering-motivated features contribute much to pronoun resolution. © 2009 ACL and AFNLP.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Semi-supervised learning for semantic relation classification using stratified sampling strategy

Semi-supervised learning for semantic relation classificatio...

引用

2009 Conference on Empirical Methods in Natural Language processing, EMNLP 2009, Held in Conjunction with ACL-IJCNLP 2009

作者： Qian, Longhua Zhou, Guodong Kong, Fang Zhu, Qiaoming Jiangsu Provincial Key Lab for Computer Information Processing Technology School of Computer Science and Technology Soochow University 1 Shizi Street Suzhou 215006 China

This paper presents a new approach to selecting the initial seed set using stratified sampling strategy in bootstrapping-based semi-supervised learning for semantic relation classification. First, the training data is partitioned into several strata according to relation types/subtypes, then relation instances are randomly sampled from each stratum to form the initial seed set. We also investigate different augmentation strategies in iteratively adding reliable instances to the lab.led set, and find that the bootstrapping procedure may stop at a reasonable point to significantly decrease the training time without degrading too much in performance. Experiments on the ACE RDC 2003 and 2004 corpora show the stratified sampling strategy contributes more than the bootstrapping procedure itself. This suggests that a proper sampling strategy is critical in semi-supervised learning. © 2009 ACL and AFNLP.

关键词： Supervised learning

来源：评论

学校读者我要写书评

暂无评论

Improving nominal SRL in Chinese language with verbal SRL information and automatic predicate recognition

Improving nominal SRL in Chinese language with verbal SRL in...

引用

2009 Conference on Empirical Methods in Natural Language processing, EMNLP 2009, Held in Conjunction with ACL-IJCNLP 2009

作者： Li, Junhui Zhou, Guodong Zhao, Hai Zhu, Qiaoming Qian, Peide Jiangsu Provincial Key Lab. for Computer Information Processing Technologies School of Computer Science and Technology Soochow University Suzhou 215006 China Department of Chinese Translation and Linguistics City University of HongKong Hong Kong

This paper explores Chinese semantic role lab.ling (SRL) for nominal predicates. Besides those widely used features in verbal SRL, various nominal SRL-specific features are first included. Then, we improve the performance of nominal SRL by integrating useful features derived from a state-of-the-art verbal SRL system. Finally, we address the issue of automatic predicate recognition, which is essential for a nominal SRL system. Evaluation on Chinese NomBank shows that our research in integrating various features derived from verbal SRL significantly improves the performance. It also shows that our nominal SRL system much outperforms the state-of-the-art ones. © 2009 ACL and AFNLP.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Robust dynamic skeleton extraction for blood vessels based on the level set method

Robust dynamic skeleton extraction for blood vessels based o...

引用

6th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2009

作者： Cui, Zhiming Xu, Jing Feng, Daming Wu, Jian Institute Of Intelligent Information Processing and Application Soochow University Jiangsu 215006 China Department Of Computer S and T Soochow University Jiangsu Province Computer IT Key Lab. 215006 China

ISBN: (纸本)9780769537351

Skeleton has very important applications in objects expression, data compression, computer vision and animation. In the discrete space, the basic skeleton algorithms have two categories: one is thinning, the other is based on the distance transformation, in a high-dimensional space generatedfrom the suiface to form the ridge to create a skeleton. The skeleton which is based on the distance transform algorithm is accurate and smooth, but we must carefully check its continuity. Such a check of the continuity would be very difficult when the skeleton structure is complex. The biggest advantage ofthinning algorithm is that it can ensure that the skeleton is in a row and keeps the main topology of the initial target, but the general position is not accurate. In this paper, combining thinning algorithms and level set model, a dynamic robust vascular skeleton extraction algorithm is proposed. First using thinning technology to generate initial skeleton in a row, and then based on the level set model, it will guide the initial framework to the correct position. In this paper, the skeleton extracted from our algorithm not only maintains the exact location and the smooth appearance in a row at the same time curbs the noise of the border with a good robustness. © 2009 IEEE.

关键词： Blood vessels

来源：评论

学校读者我要写书评

暂无评论

Robust centerline extraction for tree-like blood vessels based on the region growing algorithm and level-set method

Robust centerline extraction for tree-like blood vessels bas...

引用

6th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2009

作者： Xu, Jing Feng, Darning Wu, Jian Cui, Zhirning Institute Ofintelligent Information Processing and Application Soochow University Jiangsu 215006 China Department of Computer S and T Soochow University Jiangsu Province Computer IT Key Lab. 215006 China

ISBN: (纸本)9780769537351

In this paper, we present a tree-like blood vessels based on the region growing algorithm and level sets methodfor computing centerlines for both 2D and 3D shape analysis. Consider the tree-like vascular structure, we first use region growing technology to generate the shortest distance from the given root to all the pixels, ground on the distance can detect all the extreme nodes of branches automatically. Propagate wave-front from a point of skeleton line with the level sets, track the points of the largest curvature on the suiface of the wave-front, through solute differential equations, the skeleton can be gained. The proposed method is computationally inexpensive, not only maintain the position exactly and the shape smoothly and connectedly, and less sensitive to boundary noise. © 2009 IEEE.

关键词： Blood vessels

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：