三维草图交点识别的概率方法
|
摘要
三维草图语义描述是三维草图理解研究的基本问题之一,其中线型标注是三维草图语义描述的基本方法,三维草图交点识别又是三维草图线型标注中必不可少的工作,由于草图随意性和歧义性对草图识别造成的挑战,现有三维草图交点识别工作都不能为三维草图线型标注提供很好的支持。
本论文工作以积木世界为背景,提出一种识别三维草图交点的概率方法,该方法运用贝叶斯网表达和处理三维草图的随意性和歧义性,从草图中选取17个特征作为贝叶斯网的基元,不用分割笔划或对草图进行规整,最大限度地保留用户绘制的原始信息,避免错误的识别,为三维草图线型标注提供强有力的支持。
Semantics representation of 3D-sketch is fundamental to 3D-sketch understand-ing.3D-sketch junction recognition is an required part of line labeling, which is an es-sential approach to semantics representation of 3D-sketch. Whereas,the freedoms and ambiguities inherent in hand-drawn sketches present main challenge to sketch recog-nition research. Existing research in 3D-sketch junction recognition don't support line labeling well.
In this paper we propose a new probabilistic method to 3D-sketch junction recog-nition in the block world,which express and handle the freedoms and ambiguities with Bayesian network.17 features have been selected from 3D-sketch and as the bottom nodes of the bayesian network. It could reserve raw information in the pictures drawn by users as much as possible,without dividing stroke or standardizing the sketch. avoid-ing error recognition and offering strong support for line labeling in 3D-sketch.
本论文工作以积木世界为背景,提出一种识别三维草图交点的概率方法,该方法运用贝叶斯网表达和处理三维草图的随意性和歧义性,从草图中选取17个特征作为贝叶斯网的基元,不用分割笔划或对草图进行规整,最大限度地保留用户绘制的原始信息,避免错误的识别,为三维草图线型标注提供强有力的支持。
Semantics representation of 3D-sketch is fundamental to 3D-sketch understand-ing.3D-sketch junction recognition is an required part of line labeling, which is an es-sential approach to semantics representation of 3D-sketch. Whereas,the freedoms and ambiguities inherent in hand-drawn sketches present main challenge to sketch recog-nition research. Existing research in 3D-sketch junction recognition don't support line labeling well.
In this paper we propose a new probabilistic method to 3D-sketch junction recog-nition in the block world,which express and handle the freedoms and ambiguities with Bayesian network.17 features have been selected from 3D-sketch and as the bottom nodes of the bayesian network. It could reserve raw information in the pictures drawn by users as much as possible,without dividing stroke or standardizing the sketch. avoid-ing error recognition and offering strong support for line labeling in 3D-sketch.
引文
[1]Winston P. Artificial Intelligence. Addison Wesley Longman Publishing Company,2003
[2]Grimastead I, Martin R. Incremental line labelling for sketch input of solid models. Com-puter Graphics Forum,1996,15(2):155-166
[3]Schwikardt M. Digital clay:deriving digital models from freehand sketches. Automation in Construction,2000,9(1):107-115
[4]Lovett M F K. constructiong spatial representations of variable detail for sketch recognition. Proceedings of Proceedings of AAAI spring symposium on Control Mechanisms for Spatial Knowledge Processing in Cognitive/Intelligent Systems,2007
[5]Lovett M F K. Incremental learning of perceptual categories for open-domain sketch recogni-tion. Proceedings of Proceedings of International Joint Conference on Artificial Intelligence, 2007
[6]Cooper M. A rich discrete labeling scheme for line drawings of curved objects. IEEE Transactions on Pattern Analysis and Machine Intelligence,2008, (4):741-745
[7]Lovett M F K. Building and comparing qualitative descriptions of three-dimensional de-sign sketches. Proceedings of Proceedings of the 22nd International Qualitative Reasoning Workshop,2008
[8]Alvarado C, Davis R. Dynamically constructed Bayes nets for multi-domain sketch under-standing. Proceedings of Proceedings of the 19th International Joint Conference on Artificial Intelligence,2005.1407-1412
[9]Alvarado C, Davis R. SketchREAD:A Multi-Domain Sketch Recognition Engine. Pro-ceedings of Proceedings of the 17th annual ACM symposium on User interface software and technology. ACM,2004.23-32
[10]Rubine D. Specifying gestures by example. Proceedings of Proceedings of International Conference on Machine Learning and Cybernetics,1991
[11]Kurtoglu T. Interpreting Schematic Sketches Using Physical Reasoning. Proceedings of AAAI Spring Symposium on Sketch Understanding,2002.78-85
[12]Gennari L S T. Combining Geometry and Domain Knowledge to Interpret Hand-drawn Diagrams. Computers and Graphics,2005, (29(4))
[13]Liao S Z, Wang X J, Lu J L. An Incremental Bayesian Approach To Sketch Recognition. Proceedings of Proceedings of International Conference on Machine Learning and Cyber-netics,2005.50-63
[14]Sun Z, Zhang L, Zhang B. Online Composite Sketchy Shape Recognition Based on Bayesian Networks. Natural Computation Techniques Applications,2006,4222:506-515
[15]Zhang L S, Sun Z X. Research on Sketch Symbol Recognition Methods:[PhD Thesis]. Nanjing University,2003
[16]Alvarado C, Davis R. Resolving ambiguities to create a natural computer-based sketching environment. Proceedings of International Conference on Computer Graphics and Interac-tive Techniques, New York,2007
[17]Alvarado C. A Natural Sketching Environment:Bringing the Computer into Early Stages of Mechanical Design:[Master Thesis]. Massachusetts Institute of Technology,2000
[18]Kern R.人工智能手册.科学出版社,1991
[19]张连文.贝叶斯网引论.北京：社会科学出版社,2006
[20]Charniak E. Bayesian Networks without Tears. AI Magazine,1991,12:55-63
[21]Pearl J. Propagation and Structuring in Belief Networks. Technical Report 850022(R-42), UCLA Computer Science Department,1986
[22]Hu K B W. HMM Based On-Line Handwriting Recognition. IEEE Transaction on Pattern Analysis and Machine Intelligence,1996,18:No.10
[23]Sezgin R S T. Sketch Based Interface:Early Processing for Sketch Understanding. Proceed-ings of Proceedings of PUI,2001
[24]Sezgin R. Scale-space Based Feature Point Detection for Digital Ink. Proceedings of Inter-national Conference on Computer Graphics and Interactive Techniques,2006
[25]粟阳.笔试用户界面研究：[PhD Thesis].北京：北京科学院软件研究所,2002
[26]Lipson M. Classification of Sketch Strokes and Corner Detection using Conic Sections and Adaptive Clustering. Transactions of the ASME-R-Journal of Mechanical Design,1997.
[27]Calhoun C, Stahovich F, Kurtoglu T, et al. Recognizing Multi-Stroke Symbols. In AAAI Spring Symposium on Sketch Understanding,2002, SS-02-08:15-23
[28]Olga V. Perceptually Based Learning of Shape Descriptions for Sketch Understanding: [Master Thesis]. Washington:University of Washington,2003
[29]Hammond T, Davis R. LADDER, A Sketching Language for User Interface Developers. Proceedings of International Conference on Computer Graphics and Interactive Techniques, 2007
[30]Kevin M. Bayes Net Toolbox for Matlab/How to use the toolbox,1997-2002. http://www.cs.ubc.ca/-murphyk/Software/BNT/bnt.html
[2]Grimastead I, Martin R. Incremental line labelling for sketch input of solid models. Com-puter Graphics Forum,1996,15(2):155-166
[3]Schwikardt M. Digital clay:deriving digital models from freehand sketches. Automation in Construction,2000,9(1):107-115
[4]Lovett M F K. constructiong spatial representations of variable detail for sketch recognition. Proceedings of Proceedings of AAAI spring symposium on Control Mechanisms for Spatial Knowledge Processing in Cognitive/Intelligent Systems,2007
[5]Lovett M F K. Incremental learning of perceptual categories for open-domain sketch recogni-tion. Proceedings of Proceedings of International Joint Conference on Artificial Intelligence, 2007
[6]Cooper M. A rich discrete labeling scheme for line drawings of curved objects. IEEE Transactions on Pattern Analysis and Machine Intelligence,2008, (4):741-745
[7]Lovett M F K. Building and comparing qualitative descriptions of three-dimensional de-sign sketches. Proceedings of Proceedings of the 22nd International Qualitative Reasoning Workshop,2008
[8]Alvarado C, Davis R. Dynamically constructed Bayes nets for multi-domain sketch under-standing. Proceedings of Proceedings of the 19th International Joint Conference on Artificial Intelligence,2005.1407-1412
[9]Alvarado C, Davis R. SketchREAD:A Multi-Domain Sketch Recognition Engine. Pro-ceedings of Proceedings of the 17th annual ACM symposium on User interface software and technology. ACM,2004.23-32
[10]Rubine D. Specifying gestures by example. Proceedings of Proceedings of International Conference on Machine Learning and Cybernetics,1991
[11]Kurtoglu T. Interpreting Schematic Sketches Using Physical Reasoning. Proceedings of AAAI Spring Symposium on Sketch Understanding,2002.78-85
[12]Gennari L S T. Combining Geometry and Domain Knowledge to Interpret Hand-drawn Diagrams. Computers and Graphics,2005, (29(4))
[13]Liao S Z, Wang X J, Lu J L. An Incremental Bayesian Approach To Sketch Recognition. Proceedings of Proceedings of International Conference on Machine Learning and Cyber-netics,2005.50-63
[14]Sun Z, Zhang L, Zhang B. Online Composite Sketchy Shape Recognition Based on Bayesian Networks. Natural Computation Techniques Applications,2006,4222:506-515
[15]Zhang L S, Sun Z X. Research on Sketch Symbol Recognition Methods:[PhD Thesis]. Nanjing University,2003
[16]Alvarado C, Davis R. Resolving ambiguities to create a natural computer-based sketching environment. Proceedings of International Conference on Computer Graphics and Interac-tive Techniques, New York,2007
[17]Alvarado C. A Natural Sketching Environment:Bringing the Computer into Early Stages of Mechanical Design:[Master Thesis]. Massachusetts Institute of Technology,2000
[18]Kern R.人工智能手册.科学出版社,1991
[19]张连文.贝叶斯网引论.北京：社会科学出版社,2006
[20]Charniak E. Bayesian Networks without Tears. AI Magazine,1991,12:55-63
[21]Pearl J. Propagation and Structuring in Belief Networks. Technical Report 850022(R-42), UCLA Computer Science Department,1986
[22]Hu K B W. HMM Based On-Line Handwriting Recognition. IEEE Transaction on Pattern Analysis and Machine Intelligence,1996,18:No.10
[23]Sezgin R S T. Sketch Based Interface:Early Processing for Sketch Understanding. Proceed-ings of Proceedings of PUI,2001
[24]Sezgin R. Scale-space Based Feature Point Detection for Digital Ink. Proceedings of Inter-national Conference on Computer Graphics and Interactive Techniques,2006
[25]粟阳.笔试用户界面研究：[PhD Thesis].北京：北京科学院软件研究所,2002
[26]Lipson M. Classification of Sketch Strokes and Corner Detection using Conic Sections and Adaptive Clustering. Transactions of the ASME-R-Journal of Mechanical Design,1997.
[27]Calhoun C, Stahovich F, Kurtoglu T, et al. Recognizing Multi-Stroke Symbols. In AAAI Spring Symposium on Sketch Understanding,2002, SS-02-08:15-23
[28]Olga V. Perceptually Based Learning of Shape Descriptions for Sketch Understanding: [Master Thesis]. Washington:University of Washington,2003
[29]Hammond T, Davis R. LADDER, A Sketching Language for User Interface Developers. Proceedings of International Conference on Computer Graphics and Interactive Techniques, 2007
[30]Kevin M. Bayes Net Toolbox for Matlab/How to use the toolbox,1997-2002. http://www.cs.ubc.ca/-murphyk/Software/BNT/bnt.html
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |