Structural, Syntactic, and Statistical Pattern Recognition

Name: Structural, Syntactic, and Statistical Pattern Recognition | Joint IAPR International Workshops, S+SSPR 2022, Montreal, QC, Canada, August 26-27, 2022, Proceedings
Brand: Springer
Price: 106.99 EUR
Availability: OnlineOnly

Joint IAPR International Workshops, S+SSPR 2022, Montreal, QC, Canada, August 26-27, 2022, Proceedings

Adam Krzyzak Ching Y. Suen Andrea Torsello Nicola Nobile(Editor)

Springer (Publisher)

Published on 1. January 2023

XIII, 324 pages

E-Book

PDF with digital watermarking

System requirements

978-3-031-23028-8 (ISBN)

€106.99incl. 7% vat

System requirements

for PDF with digital watermarking

E-Book Single Licence

Available for download

Description

More details

Other editions

Content

Intro
Preface
Organization
Contents
Realization of Autoencoders by Kernel Methods
1 Introduction
2 Related Work
3 Autoencoders by Kernel Methods
3.1 Encoder and Decoder
3.2 Fundamental Mapping Without Loss
3.3 Kernelized Autoencoder
4 Comparison with Neural Networks
5 Applications
5.1 Denoising Autoencoders
5.2 Generative Autoencoders
6 Discussion
7 Conclusion
References
Maximal Independent Vertex Set Applied to Graph Pooling
1 Introduction
2 Related Work
2.1 Graph Pooling
3 Proposed Method
3.1 Maximal Independent Vertex Set (MIVS)
3.2 Adaptation of MIVS to Deep Learning
4 Experiments
4.1 Datasets
4.2 Model Architecture and Training Procedure
4.3 Ablation Studies
4.4 Comparison of MIVSPool According to Other Methods
5 Conclusion
References
Annotation-Free Keyword Spotting in Historical Vietnamese Manuscripts Using Graph Matching
1 Introduction
2 Kieu Database
3 Annotation-Free Keyword Spotting (KWS)
3.1 Synthetic Dataset Creation
3.2 Character Detection
3.3 Graph Extraction
3.4 Graph Matching
3.5 Keyword Spotting (KWS)
4 Experimental Evaluation
4.1 Task Setup and Parameter Optimization
4.2 Results
4.3 Ablation Study
5 Conclusions
References
Interactive Generalized Dirichlet Mixture Allocation Model
1 Introduction
2 Model Description
3 Variational Inference
4 Interactive Learning Algorithm
5 Experimental Results
6 Conclusion
References
Classifying Me Softly: A Novel Graph Neural Network Based on Features Soft-Alignment
1 Introduction
2 Related Work
3 Features Soft-Alignment Graph Neural Networks
4 Experiments
4.1 Experimental Setup
4.2 Ablation Study
4.3 Graph Classification Results
4.4 Graph Regression Results
5 Conclusion
References
Review of Handwriting Analysis for Predicting Personality Traits
1 Introduction
1.1 History
1.2 Applications
1.3 Requirements
2 Research Progress
2.1 Advantages
2.2 Disadvantages
3 Research Steps
3.1 Database
3.2 Pre-processing
3.3 Feature Extraction
3.4 Personality Trait
3.5 Prediction Model
3.6 Performance Measurement
4 Experiment and Future Work
4.1 Experiment
4.2 Future Work
References
Graph Reduction Neural Networks for Structural Pattern Recognition
1 Introduction and Related Work
2 Graph Matching on GNN Reduced Graphs
2.1 Graph Reduction Neural Network (GReNN)
2.2 Classification of GReNN Reduced Graphs
3 Empirical Evaluations
3.1 Datasets and Experimental Setup
3.2 Analysis of the Structure of the Reduced Graphs
3.3 Classification Results
3.4 Ablation Study
4 Conclusions and Future Work
References
Sentiment Analysis from User Reviews Using a Hybrid Generative-Discriminative HMM-SVM Approach
1 Introduction
2 Related Work
3 Hybrid Generative-Discriminative Approach with Fisher Kernels
3.1 Hidden Markov Models
3.2 Inference on Hidden States: Forward-Backward Algorithm
3.3 Fisher Kernels
4 Experiments
4.1 Problem Modeling
4.2 Datasets
4.3 Results
5 Conclusion
References
Spatio-Temporal United Memory for Video Anomaly Detection
1 Introduction
2 Related Work
2.1 Dual-Flow Structure Based on Autoencoder
2.2 Memory
3 Methodology
4 Experiments
4.1 Datasets and Evaluation Metrics
4.2 Comparison with Existing Methods
4.3 Ablation Experiments
4.4 Running Time
5 Conclusion
References
A New Preprocessing Method for Measuring Image Visual Quality Robust to Rotation and Spatial Shifts
1 Introduction
2 Proposed Preprocessing Method
3 Experimental Results
4 Conclusions
References
Learning Distances Between Graph Nodes and Edges
1 Introduction
2 Related Work
2.1 Graph Edit Distance
2.2 Learning the Edit Costs
3 Method
3.1 The Learning Method
3.2 The Algorithm
4 Practical Experiments
5 Conclusions
References
Self-supervised Out-of-Distribution Detection with Dynamic Latent Scale GAN
1 Introduction
2 Out-of-Distribution Detection with DLSGAN
3 Experiments
3.1 Experiments Settings
3.2 Experiments Results
4 Conclusion
References
A Novel Graph Kernel Based on the Wasserstein Distance and Spectral Signatures
1 Introduction
2 Related Work
3 Background
3.1 Spectral Signatures: HKS and WKS
3.2 Wasserstein Distance
4 From Spectral Signatures to Graph Kernels
5 Experiments
5.1 Experimental Setup
5.2 Sensitivity Study
5.3 Graph Classification Results
6 Conclusion
References
Discovering Respects for Visual Similarity
1 Introduction
2 Method
2.1 Dataset Selection
2.2 Image Representation
2.3 What
2.4 Why
3 Evaluations
3.1 Automatic Interpretability and Human Validation
3.2 Human Assessment of Cluster Quality
4 Conclusion
References
Graph Regression Based on Graph Autoencoders
1 Introduction
2 Related Work
2.1 Graph Embedding and Graph Regression
2.2 Autoencoders and Graph Autoencoders
2.3 Prediction of Chemical Compound Properties
3 The Method
4 Motive and Practical Application
4.1 Database
4.2 Architecture Configuration
4.3 Energy Prediction
4.4 Runtime Analysis
5 Conclusions and Future Work
References
Distributed Decision Trees
1 Introduction
2 Different Tree Architectures
2.1 Hard Decision Trees
2.2 Soft Decision Trees
2.3 Budding Trees
3 Distributed Budding Trees
4 Experiments
5 Visualization
6 Conclusions
References
A Capsule Network for Hierarchical Multi-label Image Classification
1 Introduction
2 Hierarchical Multi-label Capsules
3 Experiments
3.1 Implementation Details and Datasets
3.2 Experimental Setup
3.3 Results
4 Conclusions
References
Monte Carlo Dropout for Uncertainty Analysis and ECG Trace Image Classification
1 Introduction
2 Literature Review
2.1 ECG Classification
2.2 Uncertainty Estimation in Medical Image Analysis
3 Proposed Methodology
3.1 Dataset
3.2 Data Preprocessing
3.3 CNN Architecture
3.4 Monte Carlo Dropout
4 Experimental Results
4.1 Experimentation
4.2 Result Analysis and Discussion
5 Conclusion and Future Work
References
One-Against-All Halfplane Dichotomies
1 Introduction
2 Prior Work
3 One-Against-All Halfplane Dichotomies
4 A Geometric Perspective
5 Pairwise Attribute Difference Vectors
6 Linear Programming and Neural Networks
7 Ranking
8 Summary
References
Fast Distance Transforms in Graphs and in Gmaps
1 Introduction
1.1 Notations and Definitions
2 Distance Transform in a Graph
2.1 Geodesic Distance Transform
3 Distance Transforms in n-Gmaps
4 Results
5 Conclusions
References
Retargeted Regression Methods for Multi-label Learning
1 Introduction
2 Proposal: Retargeted Multi-label Least Square Regression
2.1 Notations
2.2 Brief Review of ReLSR
2.3 Problem Formulation
2.4 Optimization
2.5 Computational Complexity
2.6 Learning Threshold
3 Experiments
3.1 Dataset and Evaluation Measurement
3.2 Settings
3.3 Results
4 Conclusion
References
Transformer with Spatio-Temporal Representation for Video Anomaly Detection
1 Introduction
2 Related Work
2.1 Video Anomaly Detection
2.2 Transformer
3 Methodology
4 Experiments
4.1 Datasets and Evaluation Metrics
4.2 Implementation Details
4.3 Comparison with Existing State-of-the-Arts
4.4 Ablation Experiments
4.5 Running Time
5 Conclusion
References
Efficient Leave-One-Out Evaluation of Kernelized Implicit Mappings
1 Introduction
2 Kernelized Implicit Mapping
3 LOO Evaluation of KIM
4 Positioning of this Study
5 Kernels
6 Multiple Applications of LOO Matrix
6.1 Visualization and Model Selection
6.2 Nonlinear Classification
6.3 Recovery
7 Analysis of LOO Matrix
8 Discussion
9 Conclusion
References
Graph Similarity Using Tree Edit Distance
1 Introduction
2 Motivation and Basic Concepts
3 Algorithm
4 Experimental Results
5 Conclusion
References
Data Augmentation on Graphs for Table Type Classification
1 Introduction
2 Related Works
3 Method
3.1 Preprocessing
3.2 Data Augmentation
3.3 Model
4 Experiments
4.1 The Tab2Know Dataset
4.2 Using the Dataset
4.3 Results
5 Conclusions
References
Improved Training for 3D Point Cloud Classification
1 Introduction
2 Related Work
3 Method
3.1 Preliminaries
3.2 Proposed Method
3.3 Proposed Training Protocol
4 Experimental Results
4.1 Hyper-parameter Sensitivity
4.2 Results for Proposed Model Variants
4.3 Results for Explored Loss Functions
4.4 Explored Training Protocol
4.5 Effect of Augmentation
4.6 Transfer Learning
4.7 Confusion Matrix
5 Conclusions
References
On the Importance of Temporal Features in Domain Adaptation Methods for Action Recognition
1 Introduction
2 Related Works
3 Recalls Basics of an Architecture for Domain Adaptation
4 The New Designed Architecture
5 Experiments and Results
5.1 Datasets and Metrics
5.2 Parameters Setting Details
5.3 Results and Comments
6 Conclusions
References
Zero-Error Digitisation and Contextualisation of Piping and Instrumentation Diagrams Using Node Classification and Sub-graph Search
1 Introduction
2 Predicting Improperly Identified Components
2.1 Graph Representation and Data Embedding
2.2 Machine Learning and Verification
3 Searching Groups of Components
3.1 Input and Output Parameters of Our Method
3.2 Algorithm
4 Conclusions and Future Work
References
Refining AttnGAN Using Attention on Attention Network
1 Introduction
2 Background
3 Method
3.1 AttnGAN
3.2 Global Vector Attention
3.3 Attention on Attention
3.4 Mode-seeking Function
3.5 Refined-GAN Architecture
3.6 Objective Function
3.7 Implementation Details
4 Results
4.1 Datasets
4.2 Quantitative Results
4.3 Qualitative Results
5 Conclusion
References
An Autoencoding Method for Detecting Counterfeit Coins
1 Introduction
2 Selecting Similarity Criterion
2.1 Comparison of Images
2.2 Selecting a Similarity Function
3 Autoencoder Architecture
4 Training the Autoencoder
5 Custom Classifier
6 Experimental Results
6.1 Dataset
6.2 Results
7 Conclusion
References
Tarragona Graph Database for Machine Learning Based on Graphs
1 Introduction
2 Database Format
3 Database Description
3.1 Letters
3.2 Rotation Zoom
3.3 House-Hotel
3.4 Horse
3.5 Palmprint
3.6 Sagrada Familia 3D
3.7 Fingerprint
3.8 Proteins
3.9 Mutagenicity
3.10 Grec
4 Database Interaction
5 Conclusions
References
Human Description in the Wild: Description of the Scene with Ensembles of AI Models
1 Introduction
2 Related Work
3 Methods
3.1 Open Pose Estimation
3.2 Clothing Recognition by Object Detection
3.3 Clothing Color Recognition
3.4 Action Recognition
3.5 Gender and Age Classification from the Face
4 Dataset
5 Experimentation and Results
5.1 Clothing Recognition
5.2 Actions Classification
5.3 Gender and Age Classification
6 Conclusion
References
Author Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Structural, Syntactic, and Statistical Pattern Recognition

Description

More details

Other editions

Additional editions

Content

System requirements