Structural, Syntactic, and Statistical Pattern Recognition : Joint IAPR International Workshops, S+SSPR 2022, Montreal, QC, Canada, August 26-27, 2022, Proceedings / / Adam Krzyzak [and three others] (editors)

• Pubblicazione: Cham, Switzerland : , : Springer Nature Switzerland AG, , [2022]; ©2022
• Descrizione fisica: 1 online resource (336 pages)
• Disciplina: 006.37
• Soggetto topico: Computer vision; Pattern recognition systems
• Persona (resp. second.): KrzyzakAdam
• Nota di bibliografia: Includes bibliographical references and index.
• Nota di contenuto: 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.
• Titolo autorizzato: Structural, Syntactic, and Statistical Pattern Recognition
• ISBN: 3-031-23028-0
• Formato: Materiale a stampa
• Livello bibliografico: Monografia
• Lingua di pubblicazione: Inglese
• Record Nr.: 9910639901403321
• Lo trovi qui: Univ. Federico II
• Serie: Lecture notes in computer science ; ; Volume 13813.