Advances in Computer Vision and Computer Graphics (eBook)

Title page 2
Preface 6
Organization 8
Table of Contents 12
An Improved Color Mood Blending Between Images Via Fuzzy Relationship 17
Evaluation of Alzheimer’s Disease by Analysis of MR Images Using Multilayer Perceptrons, Polynomial Nets and Kohonen LVQ Classifiers 28
Joint Bayesian PET Reconstruction Algorithm Using a Quadratic Hybrid Multi-order Prior 39
Automatic Combination of Feature Descriptors for Effective 3D Shape Retrieval 52
Spatio-temporal Reflectance Sharing for Relightable 3D Video 63
Interactive Hierarchical Level of Detail Level Selection Algorithm for Point Based Rendering 75
Fast Ray-Triangle Intersection Computation Using Reconfigurable Hardware 86
An Applicable Hierarchical Clustering Algorithm for Content-Based Image Retrieval 98
MADE: A Composite Visual-Based 3D Shape Descriptor 109
Research of 3D Chinese Calligraphic Handwriting Recur System and Its Key Algorithm 121
Clouds and Atmospheric Phenomena Simulation in Real-Time 3D Graphics 133
Feature Points Detection Using Combined Character Along Principal Orientation 144
Fast Virtual Cloth Energy Minimization 155
Model-Based Feature Extraction for Gait Analysis and Recognition 166
Interactive System for Efficient Video Cartooning 177
Virtual Reality Technology Used to Develop Didactic Models 189
Copying Behaviour of Expressive Motion 196
Illumination Compensation Algorithm Using Eigenspaces Transformation for Facial Images 208
Reverse Engineering Garments 216
3D Reconstruction of a Human Face from Images Using Morphological Adaptation 228
Robust Automatic Data Decomposition Using a Modified Sparse NMF 241
A Brain MRI/SPECT Registration System Using an Adaptive Similarity Metric: Application on the Evaluation of Parkinson’s Disease 251
Hand Gesture Recognition with a Novel IR Time-of-Flight Range Camera–A Pilot Study 263
3D Reconstruction of Human Faces from Occluding Contours 277
The Multiresolution Analysis of Triangle Surface Meshes with Lifting Scheme 290
A Note on the Discrete Binary Mumford-Shah Model 299
Model-Based Plane-Segmentation Using Optical Flow and Dominant Plane 311
A Study on Eye Gaze Estimation Method Based on Cornea Model of Human Eye 323
Introduction 323
Generation of Expression Space for Realtime Facial Expression Control of 3D Avatar 334
Improving Efficiency of Density-Based Shape Descriptors for 3D Object Retrieval 346
Segmentation of Soft Shadows Based on a Daylight- and Penumbra Model 357
Sub-pixel Edge Fitting Using B-Spline 369
Re-mapping Animation Parameters Between Multiple Types of Facial Model 381
Data-Driven Animation of Crowds 393
A 3-D Mesh Sequence Coding Using the Combination of Spatial and Temporal Wavelet Analysis 405
Detection of Wilt by Analyzing Color and Stereo Vision Data of Plant 416
Human Silhouette Extraction Method Using Region Based Background Subtraction 428
Facial Feature Point Extraction Using the Adaptive Mean Shape in Active Shape Model 437
Use of Multiple Contexts for Real Time Face Identification 446
Computerized Bone Age Assessment Using DCT and LDA 456
Natural Image Matting Based on Neighbor Embedding 465
Epipolar Geometry Via Rectification of Spherical Images 477
Parallel Implementation of Elastic Grid Matching Using Cellular Neural Networks 488
Automatic Segmentation of Natural Scene Images Based on Chromatic and Achromatic Components 498
3D Model-Based Tracking of the Human Body in Monocular Gray-Level Images 510
Measurement of the Position of the Overhead Electric-Railway Line Using the Stereo Images 522
Hand Shape Recognition by Hand Shape Scaling, Weight Magnifying and Finger Geometry Comparison 532
Volumetric Bias Correction 541
Object Tracking with Particle Filter Using Color Information 550
Fitting Subdivision Surface Models to Noisy and Incomplete 3-D Data 558
Classification of Facial Expressions Using K-Nearest Neighbor Classifier 571
Cortical Bone Classification by Local Context Analysis 583
Line Segment Based Watershed Segmentation 595
A New Content-Based Image Retrieval Approach Based on Pattern Orientation Histogram 603
A Robust Eye Detection Method in Facial Region 612
Accuracy Improvement of Lung Cancer Detection Based on Spatial Statistical Analysis of Thoracic CT Scans 623
Author Index 635

Automatic Combination of Feature Descriptors for E.ective 3D Shape Retrieval (p.49)
Abstract.
We focus on improving the effectiveness of content-based 3D shape retrieval. Motivated by retrieval performance of several individual 3D model feature vectors, we propose a novel method, called prior knowledge based automatic weighted combination, to improve the retrieval effectiveness. The method dynamically determines the weighting scheme for different feature vectors based on the prior knowledge. The experimental results show that the proposed method provides significant improvements on retrieval e.ectiveness of 3D shape search with several measures on a standard 3D database. Compared with two existing combination methods, the prior knowledge weighted combination technique has gained better retrieval effectiveness.

1 Introduction

With the rapid development of 3D scanner technology, graphic hardware, and the World-Wide Web, there has been an explosion in the number of 3D models available on the Internet. In order to make use of these 3D models, the techniques of effective 3D shape retrieval become increasingly significant. 3D models can be annotated by keywords at first, facilitating the text-based retrieval. However, this is not a promising approach, because generally annotations are manually created, which is prohibitively expensive and subject to some factors.

To overcome the disadvantages of annotation-based approach, the so-called contentbased 3D shape retrieval, using the 3D model itself, has been proposed as an alternative mechanism [9]. In [17], Min compared four text annotation-based matching methods and four content-based retrieval approaches, and the experiments showed that the relatively simple solution of using only associated text for retrieval of 3D model was not as effective as using their shape.

As a promising approach applied in many fields, the content-based 3D shape retrieval has attracted many researchers in recent years. In the computer aided design [23], the similar search for standard parts is handy in helping to reach at higher speed with lower cost. In bioinformatics [11], the detection and retrieval of similar protein molecules is applied. Other cases of using this method can be found in the entertainment industry, visual reality, and so forth.

In this paper, we experimentally compare a range of di.erent 3D feature vectors, and the experimental results show that the relative ordering of feature vectors by retrieval effectiveness depends on query models or model classes, which means that no single feature vector can always outperform other feature vectors on all query models. To address the issue and improve the effectiveness of content-based 3D shape retrieval, we propose a novel method, called prior knowledge based automatic weighted combination, which provides significant improvements on retrieval effectiveness of content-based 3D shape search.

Compared with two existing methods, one is using entropy impurity, the other is based on purity-weighted, our method achieves better 3D shape retrieval performance. The rest of this paper is organized as follows. Section 2 introduces the similarity search of 3D objects about feature-based approaches and some feature vectors. Effectiveness measures and retrieval performance for single feature vectors are described in Section 3.