Search

Browse Subject Areas

For Authors

Submit a Proposal

Graph Convolutional Neural Networks for Computer Vision

Edited by Malini Alagarsamy, Rajesh Kumar Dhanaraj, J. Felicia Lilian, Vandana Sharma, and Gheorghita Ghinea
Copyright: 2025   |   Expected Pub Date:2025//
ISBN: 9781394356331  |  Hardcover  |  
298 pages
Price: $225 USD
Add To Cart

One Line Description
Revolutionize your machine learning practice with this essential book that provides expert insights into leveraging Graph Convolutional Networks (GCNNs) to overcome the limitations of traditional CNNs.

Audience
Information communication technology industrialists, academic researchers, scientists, and engineers in the fields of Artificial Intelligence, machine learning, deep learning, and data science

Description
In the last decade, computer vision has become a major focus for addressing the world’s growing processing needs. Many existing deep learning architectures for computer vision challenges are based on convolutional neural networks (CNNs). Despite their great achievements, CNNs struggle to encode the intrinsic graph patterns in specific learning tasks. In contrast, graph convolutional networks have been used to address several computer vision issues with equivalent or superior results. The use of GCNNs has shown significant achievement in image classifications, video understanding, point clouds, meshes, and other applications in natural language processing. This book focuses on the applications of graph convolutional networks in computer vision. Through expert insights, it explores how researchers are finding ways to perform convolution algorithms on graphs to improve the way we use machine learning.

Back to Top
Author / Editor Details
Malini Alagarsamy, PhD is an assistant professor at the Thiagarajar College of Engineering. She has published more than 30 research papers in journals and national and international conferences. Her research interests include software engineering, mobile application development, green computing, Internet of Things, blockchain, and machine learning.

Rajesh Kumar Dhanaraj, PhD is a Professor in the School of Computing Science and Engineering at Galgotias University. He has authored and edited more than 25 books and 53 articles in international journals and conferences and holds 21 patents. His research interests include machine learning, cyber-physical systems, and wireless sensor networks.

J. Felicia Lilian is an Assistant Professor at the Thiagarajar College of Engineering. She has published more than 10 articles in international journals and conferences. Her research interests include natural language processing, machine learning, and deep learning.

Vandana Sharma, PhD is an Associate Professor at the Amity Institute of Information Technology at the Amity University Noida Campus with more than 14 years of teaching experience. She has published 25 research papers in international journals and conferences. Her primary areas of interest include artificial intelligence, machine learning, blockchain technology, and the Internet of Things (IoT).

George Ghinea, PhD is a Professor in the Department of Computer Science at Brunel University London. He has more than 600 publications to his credit, including book chapters and research articles in international journals of repute. His research centers on extending the notion of multimedia with that of mulsemedia, a term to denote multiple sensorial media.

Back to Top

Table of Contents
Preface
1. Role of Graph Convolutional Neural Networks (GCNN) in Computer Vision Applications
A. Malini, Vandana Sharma, J. Felicia Lilian, Rajesh Kumar Dhanaraj, Sharangapriyan S. and Shrinivas S.
1.1 Introduction
1.2 Understanding Convolutional Neural Network in Computer Vision
1.3 Core Components of CNN
1.3.1 Hierarchy Feature Learning
1.4 Extending CNNs to Handle Graph-Structured Data
1.4.1 Graphs—A Universal Data Structure
1.4.2 Challenges in Processing Graphs with CNN
1.4.3 Graph Convolutional Neural Networks (GCNNs): Bridging the Gap
1.4.4 Architectural Components of Graph Convolutional Layers
1.4.5 Graph Convolutional Layers: Adaptation of Convolutional Layers
1.5 Application of GCNN in Computer Vision
1.5.1 Semantic Segmentation
1.5.2 Object Detection and Localization with GCNN
1.5.3 Graph-Based Image Classification
1.5.4 Point Cloud Analysis and 3D Object Recognition
1.5.5 Video Understanding and Action Recognition through Graphs
1.6 Enhancing Performance and Interpretability with GCNN
1.6.1 Handling Graph Irregularities and Noisy Data
1.6.2 Transfer Learning and Pre-Trained GCNN Models
1.6.2.1 Transfer Learning Model for Waste Classification
1.6.2.2 Pre-Trained GCNN Models to Identify Finger Vein
1.6.3 Addressing Overfitting and Generalization Challenges
1.7 Future Directions and Emerging Trends
1.7.1 Advances in Graph Neural Network Architectures
1.7.1.1 Graph Attention Network
1.7.1.2 GNNExplainer: Generating Explanations for Graph Neural Networks
1.7.2 Integration of Graph Structures with Traditional CNNs
1.7.2.1 Locally Connected and Spectral Networks on Graphs
1.7.3 Explainable AI in GCNNs for Computer Vision
1.8 Challenges and Open Research Questions
1.8.1 Scalability of GCNNs to Large Graphs
1.8.1.1 Scalable Graph Convolutional Networks
1.8.2 Robustness of GCNNs Against Adversarial Attacks
1.8.2.1 Understanding Adversarial Robustness of Symmetric Network
1.8.3 Combining Graph-Based and Spatial Feature Representations
1.8.3.1 Analysis of Graph Convolutional Networks and Recent Datasets for Visual Question Answering
1.8.4 Ethical Considerations and Fairness in GCNN Applications
1.8.4.1 Neutrality and Abstraction in Sociotechnical Systems
1.9 Case Studies: Real-World Applications
1.10 Conclusion
References
2. Scene Graph Generation from Static Images: Overview, Methods, and Applications
K. Krishnakishore, R. Vijayarangan, V. Jagan Naveen and V. Kannan
2.1 Introduction
2.2 Definition
2.3 Challenge
2.4 Scene Graph Generation
2.5 Static Image
2.6 Degradation of a Static Image
2.6.1 Blurring Model
2.6.2 Affine Transformation
2.7 Method 1: Wavelet Feature Extraction
2.8 Psychological Perspective
2.9 Linguistic Perspective
2.10 Concepts and Conceptual Structures in Artificial Intelligence Perspective
2.11 Applications of CGS
2.12 Linguistic and Psychological Perspective
2.13 Image Synthesis from Layouts
2.14 Method Comparison
2.15 Conclusion
References
3. Transformation from CNN to Graph-Structured Data: Node Classification and Edge Prediction
R. Vijayarangan, R. Satish Kumar, K. Umadevi and K. Ashok Kumar
3.1 Why Graphs
3.1.1 The Definition of “Graph Convolution”
3.1.2 Graph Fourier Transform
3.1.3 Developing a Comprehensive Neural Network
3.2 SVM (Support Vector Machine)
3.3 XGBOOST
3.4 Artificial Neural Network (ANN)
3.5 Auto Encoder (AE)
3.6 Demographic and Related Data: Health Condition, Type of Gender, Age, Family Condition
3.7 Naïve Bayes (NB)
3.7.1 K-Star
3.8 Random Forest (RF)
3.8.1 Logistic Regression (LR)
3.8.2 C4.5 Algorithm
3.9 Conclusions
References
4. Research Trends and Challenges of GCNN Over CNN and Digital Image Processing Techniques
Rithish Kanna S., Suganthi P. and Kavitha P.
4.1 Introduction
4.1.1 Frequently Used Artistic Styles
4.2 Introduction to Convolutional Neural Network
4.2.1 Types of CNN
4.2.1.1 1D CNN
4.2.1.2 2D CNN
4.2.1.3 3D CNN
4.2.2 Application of CNN
4.2.2.1 Face Detection
4.2.2.2 Object Detection
4.2.2.3 X-Ray Image Analysis
4.2.2.4 Other Applications
4.3 Neural Style Transfer—Artistic View
4.4 Various Existing Works of NST
4.4.1 Literature Survey
4.5 Hybrid Neural Style Transfer
4.5.1 Scenarios Considered
4.5.2 Algorithm of HNST
4.6 Implementation of HNST
4.6.1 OpenCV
4.6.2 Pillow
4.6.3 Imageio
4.6.4 TensorFlow
4.7 Results and Inference
4.7.1 For Sketch Images
4.7.2 For Photographic Image
4.8 Further Ideas of HNST
4.8.1 Introduction to HNST 2
4.9 Conclusion
References
5. Classification of Graph Filtering Operations and Inductive Learning by Exploiting Multiple Graphs in GCNN
S. Kayalvizhi, Harish Sekar and Prasanna Guptha M.P.
5.1 Introduction
5.2 Graph Basics
5.3 Graph Convolutional Filters
5.4 Graph Filter Banks
5.5 Graph Neural Networks
5.6 Conclusion
References
6. GCNN with Adaptive Filters for Hyperspectral Image Classification
U. Moulali, R. Vijayarangan, S. Khaleel Ahamed and Kamakshaiah Kolli
6.1 Introduction
6.2 Related Works
6.3 Classification of Graph Filtering Operations
6.3.1 Introduction to Graph Convolutional Neural Networks (GCNNs)
6.3.2 Classification of Graph Filtering Operations
6.3.2.1 Spatial Domain Filtering
6.3.2.2 Frequency Domain Filtering
6.3.2.3 Time Domain Filtering
6.3.2.4 Graph Neural Networks (GNNs)
6.3.2.5 Hybrid Filtering
6.3.2.6 Attention Mechanisms
6.3.2.7 Graph Pooling
6.3.2.8 Non-Local Filtering
6.4 Experimental Analysis and Discussion
6.5 Conclusion
References
7. Graph Convolution Neural Network on Human Motion Prediction
B. Subbulakshmi, M. Nirmala Devi and Srimadhi J.
7.1 Introduction
7.1.1 Human Motion Prediction
7.1.2 Traditional Methods
7.1.3 Deep Learning Methodologies
7.2 Graph Convolution Neural Network (GCN)
7.2.1 Why Graph Convolution Network?
7.3 Forms of GCN on Human Motion Prediction
7.3.1 Gated Graph Convolutional Network
7.3.2 Spatial and Temporal Graph Convolutional Network
7.3.3 Dynamic Multi-Scale Spatiotemporal Graph Convolution Network
7.3.4 Multi-Scale Residual Graph Convolution Network (MSR-GCN)
7.4 Types of Graphs Employed on GCN
7.4.1 Skeleton-Based Pose as an Undirected Graph
7.4.2 Connective Graph
7.4.3 Global Graph
7.5 Conclusion
References
8. GraphChXNet: A Graph Convolutional Neural Network-Based Model for Detecting Chest Diseases Using X-Ray Images
D. Kiruthika, N. Vinothini, G. Jegan and G. Ananthi
8.1 Introduction
8.2 Proposed Methodology
8.3 Results and Discussion
8.4 Conclusion
References
9. Aspect-Based Sentiment Analysis Using GCN
Sachin K., Santhosh K.M.R., Sugindar A.D. and Dr. J. Felicia Lilian
9.1 Introduction
9.2 GCN and ABSA
9.2.1 The Meaning of ABSA
9.2.2 Meaning of GCN
9.3 Advancements of GCN and ABSA over the Years
9.3.1 Early Stages of ABSA
9.3.2 Establishment of Various GCN Models into ABSA
9.3.3 Table Consisting of Details of ABSA and Technology They Were Built Upon
9.4 Advancement of Technology with GCN and Algorithm Used
9.4.1 Algorithm
9.4.2 Advancement of Technology Using GCN
9.5 Case Study on GCN Application: Recommendation Systems
9.5.1 Before Arrival of GCN in the Field of Recommendation Systems
9.5.2 After Arrival of GCN in the Field of Recommendation Systems
9.6 Summary
References
10. Analysis and Classification Using Graph Convolutional Neural Networks in Medical Imaging
M. Suguna and Priya Thiagarajan
10.1 Introduction
10.1.1 Artificial Intelligence (AI) Applications in Healthcare
10.1.2 Artificial Intelligence for Image Processing Tasks in Healthcare
10.1.3 Concept of Neural Networks
10.1.4 Benefits of Neural Networks
10.2 Literature Review—GCNN in Healthcare
10.3 Methodology
10.3.1 Dataset
10.3.2 Exploratory Data Analysis (EDA)
10.3.3 Data Preprocessing and Image Augmentation
10.3.4 Image Classification Using Graph Convolutional Neural Networks
10.3.4.1 Convolutional Neural Network (CNN)
10.3.4.2 Graph Convolutional Neural Network (GCNN)
10.3.5 Performance Metrics of the GCNN Image Classification System
10.4 Results and Discussion
10.5 Conclusion
References
11. Case Studies and Real-World Applications of Graph Convolutional Networks in Computer Vision
Yogeesh N.
11.1 Introduction
11.1.1 Overview of the Chapter’s Focus
11.1.2 Case Studies’ Importance, Relevance, and Applications in Real-World Situations
11.1.3 Brief Summary of the Following Sections
11.2 Graph Convolutional Networks: A Brief Review
11.2.1 Concepts of Graph and Convolutional Neural Networks
11.2.2 Recap of Key Concepts and Principles of Graph Convolutional Networks (GCNs)
11.2.3 Node Classification of Graph Convolutional Networks
11.2.4 Graph Convolutional Networks for Graph Classification
11.2.5 Graph Convolutional Networks for Link Prediction
11.3 Case Study 1: Graph Convolutional Networks for Image Classification
11.3.1 Description of the Problem Statement and Dataset Used
11.3.2 Explanation of the Architecture and Components of the GCN Model
11.3.3 Presentation of the Experimental Results and Analysis
11.3.4 Discussion on the Strengths and Limitations of Using GCNs for Image Classification
11.4 Case Study 2: Object Detection and Localization Using Graph Convolutional Networks
11.4.1 Object Detection and Localization Overview of the Problem
11.4.2 A Description of How GCNs Can Be Used to Solve It
11.4.3 Detailed Description of the GCN-Based Object Detection
11.4.4 Experimental Results and Evaluation Metrics
11.4.5 An Analysis of the Performance and Potential Difficulties in Object Detection with GCNs
11.5 Case Study 3: Semantic Segmentation with Graph Convolutional Networks
11.5.1 Introduction to Semantic Segmentation and Its Application in Computer Vision
11.5.2 Applying GCNs to Semantic Segmentation Problems
11.5.3 Architecture of the GCN-Based Semantic Segmentation Network
11.5.4 Results and their Comparison with Traditional Segmentation Techniques
11.5.5 The Advantages and Disadvantages of GCNs in Semantic Segmentation
11.6 Case Study 4: 3D Vision and Point Cloud Processing of Graph Convolutional Networks
11.6.1 3D Vision and Point Cloud Processing Problems: An Introduction
11.6.2 How GCNs Can Be Applied to the Process of Processing 3D Data
11.6.3 Application of a GCN-Based Method for Open-Ended Triangular Rasterization Tasks
11.6.4 Experimental Results and Performance Evaluation Presentations
11.6.5 The Application of and Future Prospects for GCNs in 3D Vision
11.7 Case Study 5: Graph Convolutional Networks for Video Understanding and Action Recognition
11.7.1 Understanding and Action Recognition in Video
11.7.2 GCNs Applied to the Modeling of Spatiotemporal Relationships
11.7.3 A Revised GCN-Based Framework for Action Recognition
11.7.4 Experimental Results and Comparative Analysis
11.7.5 Examination of the Challenges and Possibilities of GCNs in Video Understanding
11.8 Other Notable Case Studies and Applications
11.8.1 Particular Case Study: Recognition of Video Action
11.8.2 Additional Case Studies and their Contributions in Brief
11.8.3 Emphasizing Different Applications Such as Scene Understanding or Medical Imaging
11.8.4 The Findings and Conclusions of these Case Studies
11.9 Discussion and Future Directions
11.9.1 Whole Case Studies and their Influence on Computer Vision
11.9.2 An Evaluation of the Major Conclusions and Trends that Emerged Across the Case Studies
11.9.3 A Discussion on the Limitations and Difficulties with a List of Improvements
11.9.4 Examination of New Ideas in Research and Future Trends
11.10 Conclusion
11.10.1 Summaries of the Cases Presented in This Chapter
11.10.2 The Importance of GCNs for Computer Vision Applications is Emphasized
11.10.3 Inspiration for Further Research and Exploration in this Area
Bibliography
12. Case Study and Use Cases of Dynamic Graphs in GCNN for Computer Vision
S. Anubha Pearline and S. Geetha
12.1 Introduction
12.1.1 Handcrafted Features
12.1.2 CNN
12.2 Graph Convolutional Neural Networks (GCNNs)
12.2.1 Graph and Images
12.2.1.1 Graph Nodes
12.2.1.2 Graph Edges
12.2.1.3 Message Passing
12.2.1.4 Aggregation Function
12.2.1.5 Graph Embeddings
12.2.1.6 Neural Network Layers
12.2.1.7 Output
12.2.2 Types of Graph Convolutional Neural Networks
12.2.2.1 Spectral Graph Convolutional Network
12.2.2.2 GraphSAGE
12.2.2.3 Multiscale Dynamic Graph Convolutional Network (MDGCN)
12.2.2.4 Spectral–Spatial Graph Convolutional Networks (S2-GCNs)
12.2.2.5 Graph Attention Networks (GATs)
12.2.2.6 Gated Graph Convolutional Neural Networks
12.2.2.7 Graph AutoEncoders (GAEs)
12.3 GCNN Case Studies
12.3.1 Image Classification
12.3.1.1 Hyperspectral Image Classification (HSI)
12.3.1.2 Breast Cancer Detection
12.3.1.3 Coronary Heart Disease Prediction
12.3.2 Object Detection
12.3.3 Image Segmentation and Scene Understanding
12.4 Challenges and Issues in GCNN for CV
12.5 Conclusion
References
About the Editors
Index

Back to Top



Description
Author/Editor Details
Table of Contents
Bookmark this page