![Cover image for Intelligent scene modelling information systems Cover image for Intelligent scene modelling information systems](/client/assets/5.0.0/ctx//client/images/no_image.png)
Available:*
Library | Item Barcode | Call Number | Material Type | Item Category 1 | Status |
---|---|---|---|---|---|
Searching... | 30000010206003 | T385 I576 2009 | Open Access Book | Book | Searching... |
On Order
Summary
Summary
Scene modeling is a very important part in Computer Graphics because it allows c- ating more or less complex models to be rendered, coming from the real world or from the designer's imagination. However, scene modeling is a very difficult task, as there is a need of more and more complex scenes and traditional geometric modelers are not well adapted to computer aided design. Even if traditional scene modelers offer very interesting tools to facilitate the designer's work, they suffer from a very important drawback, the lack of flexibility, which does not authorize the designer to use incomplete or imprecise descriptions, in order to express his (her) mental image of the scene to be designed. Thus, with most of the current geometric modelers the user must have a quite precise idea of the scene to design before using the modeler to achieve the modeling task. This kind of design is not really a computer aided one, because the main creative ideas have been elaborated without any help of the modeler. Declarative scene modeling could be an interesting alternative to traditional g- metric modeling. Indeed, declarative scene modeling tries to give intuitive solutions to the scene modeling problem by using Artificial Intelligence techniques which allow the user to describe high level properties of a scene and the modeler to give all the solutions corresponding to imprecise properties.
Table of Contents
1 Intelligent Scene Modelling Information Systems: The Case of Declarative Design Support | p. 1 |
1.1 Introduction | p. 1 |
1.2 The Scene Modelling Process in Declarative Design Support | p. 3 |
1.3 Information, Knowledge and Scene Models Representations | p. 9 |
1.3.1 Physical Scene Models | p. 12 |
1.3.2 Conceptual Scene Models - Generic Models | p. 14 |
1.3.3 Scene Conceptual Modelling in MultiCAD | p. 16 |
1.4 Software Architectures for Declarative Design Support | p. 21 |
1.4.1 MultiCAD: Objectives, Constraints and Functional Choices | p. 22 |
1.4.2 Definition of MultiCAD Framework-Architecture | p. 23 |
1.5 Conclusion | p. 25 |
References | p. 26 |
2 Declarative Modeling in Computer Graphics | p. 29 |
2.1 Introduction | p. 29 |
2.2 What Is Declarative Scene Modeling | p. 30 |
2.3 Imprecision Management in Declarative Scene Modelers | p. 31 |
2.4 A Classification of Declarative Scene Modelers | p. 32 |
2.4.1 Modelers Using Exploration Mode in Scene Generation | p. 32 |
2.4.2 Modelers Using Solution Search Mode in Scene Generation | p. 35 |
2.4.3 Other Declarative or Declarative-Like Modelers | p. 38 |
2.5 Scene Understanding in Declarative Scene Modeling | p. 40 |
2.6 Constraint Satisfaction Techniques for Declarative Scene Modeing | p. 41 |
2.6.1 Arithmetic Constraint Satisfaction Techniques | p. 41 |
2.6.1.1 The Resolution Process | p. 41 |
2.6.1.2 Constraint Logic Programming on Finite Domains - CLP(FD) | p. 42 |
2.6.1.3 Hierarchical Decomposition-Based Improvements | p. 43 |
2.6.2 Geometric Constraint Satisfaction Techniques | p. 44 |
2.6.2.1 Principles of the MultiFormes Geometric Constraint Solver | p. 45 |
2.6.2.2 The Resolution Process | p. 45 |
2.6.2.3 The Intersection and Sampling Problems | p. 45 |
2.6.2.4 Some Other Problems | p. 46 |
2.6.3 Discussion | p. 47 |
2.6.3.1 Arithmetic CSP | p. 47 |
2.6.3.2 Geometric CSP | p. 48 |
2.7 Declarative Scene Modeling and Machine-Learning Techniques | p. 48 |
2.7.1 A Dynamical Neural Network for Filtering Unsatisfactory Solutions in DMHD | p. 49 |
2.7.1.1 Structure of the Used Network | p. 49 |
2.7.1.2 The Machine Learning Process | p. 51 |
2.7.1.3 Discussion | p. 52 |
2.8 Advantages and Drawbacks of Declarative Scene Modeling | p. 53 |
2.9 Future Issues | p. 54 |
2.10 Conclusion | p. 55 |
References | p. 55 |
3 Understanding Scenes | p. 59 |
3.1 Introduction to Reverse Engineering | p. 59 |
3.1.1 Reverse Engineering in Scene Modelling | p. 60 |
3.1.2 Reverse Engineering and Geometric Modelling | p. 62 |
3.1.3 Reverse Engineering and Feature-Based Modelling | p. 63 |
3.1.4 Reverse Engineering and Declarative Modelling | p. 65 |
3.2 Integration of the Two Models | p. 68 |
3.3 Reconstruction Phase | p. 69 |
3.4 Extended Design Methodology | p. 70 |
3.5 System Architecture | p. 71 |
3.5.1 Data and Knowledge Storage | p. 73 |
3.5.2 The Stratified Representation | p. 74 |
3.5.3 Extraction of Relations and Properties | p. 77 |
3.5.4 Scene Modifications | p. 78 |
3.5.5 The Propagation Policy | p. 79 |
3.5.6 The Unified Stratified Representation | p. 81 |
3.5.7 The Resultant Declarative Description | p. 82 |
3.6 Conclusions | p. 83 |
References | p. 85 |
4 Intelligent Personalization in a Scene Modeling Environment | p. 89 |
4.1 Introduction | p. 89 |
4.2 Intelligent Personalization and Contributing Fields | p. 90 |
4.3 Preference Model | p. 92 |
4.3.1 Preference Structure | p. 92 |
4.3.2 User Preference as a Function | p. 94 |
4.4 Multicriteria Decision Support | p. 95 |
4.4.1 Outranking Methods | p. 96 |
4.4.2 Weighted Sum Methodologies | p. 97 |
4.5 Machine Learning | p. 97 |
4.5.1 Traditional Machine Learning Mechanisms | p. 98 |
4.5.2 Incremental Learning | p. 99 |
4.5.3 Imbalanced Datasets | p. 100 |
4.5.4 Context Specific Issues | p. 101 |
4.6 Intelligent Personalization in a Scene Modeling Environment | p. 102 |
4.6.1 Scene Representations | p. 102 |
4.6.2 Scene Modeling Process | p. 104 |
4.6.2.1 Solution Generation: Constraint Satisfaction Techniques | p. 104 |
4.6.2.2 Solution Generation: Evolutionary Techniques | p. 105 |
4.6.2.3 Solution Visualization | p. 105 |
4.6.2.4 Scene Modeling Environment | p. 106 |
4.6.3 Preferences Acquisition | p. 106 |
4.6.3.1 Solution Encoding for Preferences Acquisition | p. 107 |
4.6.3.2 Preferences Acquisition via Incremental Learning | p. 107 |
4.6.3.3 User-Assisted Acquisition of Preferences | p. 108 |
4.7 Intelligent User Profile Module Architecture | p. 110 |
4.7.1 Declarative Modeling | p. 110 |
4.7.2 Module Architecture | p. 112 |
4.8 Experimental Results | p. 114 |
4.8.1 Performance Indices and Representative Scenes | p. 114 |
4.8.2 Experiment Series | p. 115 |
4.9 Conclusion | p. 117 |
References | p. 117 |
5 Web-Based Collaborative System for Scene Modelling | p. 121 |
5.1 Introduction | p. 121 |
5.1.1 Research Scope | p. 123 |
5.2 Related Work | p. 124 |
5.2.1 Collaborative Design | p. 124 |
5.2.1.1 Collaborative Systems | p. 124 |
5.2.2 Declarative Design | p. 126 |
5.2.3 Overview of MultiCAD Architecture | p. 127 |
5.2.4 DKABM Framework | p. 128 |
5.2.5 Declarative Design Representations | p. 128 |
5.2.6 Collaborative Declarative Modelling System | p. 129 |
5.3 Web-Based CDMS Framework | p. 129 |
5.3.1 Declarative Collaborative Module | p. 130 |
5.4 Case Study | p. 137 |
5.4.1 Study of Collaborative Activity | p. 137 |
5.5 Team Profile Module | p. 138 |
5.5.1 Single Designer Approach | p. 138 |
5.5.1.1 Intelligent Profile Estimation | p. 139 |
5.5.1.2 Recursive Implementation | p. 140 |
5.5.2 Collaborative Approach | p. 142 |
5.5.2.1 Preference Consensus Module | p. 142 |
5.5.2.2 Collaborative Clustering | p. 144 |
5.5.3 Simulations | p. 146 |
5.6 Conclusions | p. 147 |
References | p. 148 |
6 Aesthetic - Aided Intelligent 3D Scene Synthesis | p. 153 |
6.1 Introduction | p. 153 |
6.1.1 Research Scope | p. 154 |
6.1.2 Proposed Methodology - Contributing Areas | p. 154 |
6.2 Related Work | p. 155 |
6.2.1 Evolutionary Computing Techniques | p. 156 |
6.2.1.1 Evolutionary Design | p. 157 |
6.2.1.2 Genetic Algorithm Applications in Design | p. 157 |
6.2.2 Computational Aesthetic Approaches | p. 159 |
6.2.3 Style Modelling Approaches | p. 160 |
6.2.3.1 The Concept of Style | p. 160 |
6.2.4 MultiCAD Framework Style | p. 162 |
6.3 Research Approach | p. 162 |
6.3.1 Architectural Style Modelling | p. 163 |
6.3.1.1 Style Knowledge Framework | p. 163 |
6.3.1.2 Measure of Style | p. 169 |
6.3.2 Multi-objective Genetic Algorithm | p. 169 |
6.3.2.1 Genetic Algorithm | p. 170 |
6.3.2.2 MOGA Mechanism | p. 171 |
6.4 Implementation Framework | p. 172 |
6.4.1 Software Architecture | p. 172 |
6.4.1.1 User Interface Layer | p. 173 |
6.4.1.2 Processing Layer | p. 173 |
6.4.1.3 Data Management Layer | p. 174 |
6.5 System Evaluation | p. 174 |
6.6 Discussion | p. 178 |
6.7 Conclusions | p. 179 |
6.7.1 Declarative Modelling and Architectural Conceptual Design | p. 179 |
6.7.2 Aesthetic and Artificial Intelligence | p. 180 |
References | p. 180 |
7 Network Security Surveillance Aid Using Intelligent Visualization for Knowledge Extraction and Decision Making | p. 185 |
7.1 Introduction | p. 185 |
7.1.1 Web Security | p. 186 |
7.1.2 Intrusion Detection | p. 186 |
7.1.3 Visualization | p. 187 |
7.1.4 Visual Data Analysis | p. 188 |
7.1.5 Research Objectives | p. 189 |
7.2 Related Work | p. 191 |
7.3 Visualization Prototype System | p. 192 |
7.3.1 Data Capture Module | p. 193 |
7.3.2 Pre-processor Module | p. 194 |
7.3.3 Knowledge Base Module | p. 194 |
7.3.3.1 Classes of Web Attacks | p. 194 |
7.3.3.2 Training Data Quality | p. 196 |
7.3.3.3 Evolutionary Artificial Neural Network | p. 196 |
7.3.3.4 EANN Performance Versus ANN | p. 199 |
7.3.4 Graph Generator Module | p. 200 |
7.3.5 Statistical Analysis Module | p. 205 |
7.4 Prototype System Performance | p. 206 |
7.4.1 Introduction | p. 206 |
7.4.2 Classification | p. 206 |
7.4.2.1 Neyman-Pearson Decision Rule | p. 206 |
7.4.2.2 Sufficient Statistics and Monotonic Transformations | p. 207 |
7.4.2.3 Neyman-Pearson Lemma: General Case | p. 208 |
7.4.3 Detection, False and Miss Probabilities of the Prototype System | p. 208 |
7.4.4 ROC Curve of the Prototype System | p. 210 |
7.5 Conclusion | p. 212 |
References | p. 213 |
Author Index | p. 215 |