Pattern Recognition. ICPR International Workshops and Challenges

Name: Pattern Recognition. ICPR International Workshops and Challenges | Virtual Event, January 10-15, 2021, Proceedings, Part IV
Brand: Springer
Price: 96.29 EUR
Availability: OnlineOnly

Virtual Event, January 10-15, 2021, Proceedings, Part IV

Alberto Del Bimbo Rita Cucchiara Stan Sclaroff Giovanni Maria Farinella Tao Mei Marco Bertini Hugo Jair Escalante Roberto Vezzani(Herausgeber*in)

Springer (Verlag)

Erschienen am 4. März 2021

XIX, 735 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-030-68799-1 (ISBN)

96,29 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Intro
Foreword by General Chairs
Preface
Challenges
ICPR Organization
Contents - Part IV
FGVRID - Fine-Grained Visual Recognition and re-Identification
Workshop on Fine-Grained Visual Recognition and Re-identification
Workshop Description
Organization
General Chairs
Program Committee
Densely Annotated Photorealistic Virtual Dataset Generation for Abnormal Event Detection
1 Introduction
2 Related Work
2.1 Related Tools for Synthetic Data Generation
2.2 Related Synthetic Data
2.3 Real-World Datasets for Abnormal Crowd Event Detection
3 Dataset Creation and Description
3.1 Dataset Creation: Interacting with the Virtual World
3.2 Dataset Description
4 Unsupervised Abnormal Event Detection Using GTA V
4.1 Cumulative Sum Method for Abnormal Event Detection
4.2 Experimental Results: Abnormal Event Detection
5 Conclusions
References
Unsupervised Domain Adaptive Re-Identification with Feature Adversarial Learning and Self-similarity Clustering
1 Introduction
2 Related Work
2.1 Person and Vehicle Re-ID
2.2 Unsupervised Domain Adaptation
2.3 Unsupervised Domain Adaptation Methods for Re-ID
3 Methodology
3.1 Problem Definition
3.2 Feature Extractor Pre-training
3.3 Domain-Invariant Feature Adversarial Learning
3.4 Self-similarity Clustering
3.5 Relabeling Algorithm
3.6 Training Procedure
4 Experiment
4.1 Datasets
4.2 Implementation Details
4.3 Ablation Study
4.4 Comparison with State of the Art
5 Conclusion
References
A Framework for Jointly Training GAN with Person Re-Identification Model
1 Introduction
2 Related Work
2.1 Generative Adversarial Networks
2.2 Person Re-Identification
2.3 GANs for Person Re-Identification
3 Proposed Joint Training Framework
3.1 Joint Training with Identification Loss
3.2 Joint Training with Triplet Loss
4 Experimental Results
4.1 Implementation Details
4.2 Ablation Study
4.3 Comparison to State of the Art
5 Conclusion
References
Interpretable Attention Guided Network for Fine-Grained Visual Classification
1 Introduction
2 Related Work
2.1 Fine-Grained Classification
2.2 Interpretable Neural Networks
3 Method
3.1 Data Augmentation Method
3.2 Interpretable Attention
3.3 Progressive Training Mechanism
4 Experiments
4.1 Implementation Details
4.2 Performance Comparison
4.3 Visualization
5 Conclusion
References
Use of Frequency Domain for Complexity Reduction of Convolutional Neural Networks
1 Introduction
2 Literature Review
3 CNN Computing Based on FFT
4 FFT Based Computation Using Splitting
4.1 Proposed Method
4.2 Computational Complexity and Memory Access Analysis
5 Experimental Results
6 Conclusion
References
From Coarse to Fine: Hierarchical Structure-Aware Video Summarization
1 Introduction
2 Related Work
3 Proposed Approach
3.1 Hierarchical Summarization Network
3.2 Reward Function
4 Experiments
4.1 Datasets
4.2 Video Summarization
5 Conclusion
References
ADNet: Temporal Anomaly Detection in Surveillance Videos
1 Introduction
2 Related Works
3 Method
3.1 Anomaly Detection Network
3.2 Temporal Sliding Window
3.3 Loss
3.4 Implementation Details
4 Experiments
4.1 Evaluation Metrics
4.2 Dataset
4.3 Results
5 Conclusion
References
Soft Pseudo-labeling Semi-Supervised Learning Applied to Fine-Grained Visual Classification
1 Introduction
2 Related Work
2.1 Fine-Grained Visual Classification
2.2 Semi-Supervised Learning
2.3 Semi-Supervised FGVC
3 Problem Formulation
3.1 Pseudo-label
3.2 Entropy Minimization
4 Experimental Results
4.1 Datasets
4.2 Comparison with SSL Methods
References
HCAU 2020 - The First International Workshop on Deep Learning for Human-Centric Activity Understanding
Workshop on Deep Learning for Human-Centric Activity Understanding
Workshop Description
Organization
Program Chairs
Program Committee
Spot What Matters: Learning Context Using Graph Convolutional Networks for Weakly-Supervised Action Detection
1 Introduction
2 Related Work
2.1 Action Recognition and Detection
2.2 Visual Relational Reasoning
3 Learning Context with GCN
3.1 Feature Extraction with Weak Supervision
3.2 Graph Convolutional Networks
3.3 Implementation Details
4 Experiments
4.1 Dataset and Evaluation Metric
4.2 Evaluation of Architecture Choices
4.3 Comparison with Baseline and State of the Art
4.4 Reducing Annotation to One Bounding Box
5 Analysis of Attention
5.1 Evaluation of Attention Maps
6 Conclusion
References
Social Modeling Meets Virtual Reality: An Immersive Implication
1 Introduction
2 Crowd Analysis
3 Crowd and Virtual Reality
4 Experiemental Evaluation
5 Conclusion
References
Pickpocketing Recognition in Still Images
1 Introduction
2 Background
3 Methodology
3.1 Dataset Preparation
3.2 Design
4 Experiments
5 Conclusion
References
t-EVA: Time-Efficient t-SNE Video Annotation
1 Introduction
2 Related Work
3 t-EVA for Efficient Video Annotation
3.1 How to Annotate?
4 Experiments
4.1 Datasets
4.2 Evaluation Metrics
4.3 Implementation Details
4.4 Results on ActivityNet
4.5 Generalization
5 Ablation Study
5.1 Dimensionality Reduction
5.2 t-SNE Parameters
5.3 2D-3D Comparison
6 Conclusion
References
Vision-Based Fall Detection Using Body Geometry
1 Introduction
2 Related Work
3 Proposed Method
3.1 Down-Sampling the Videos
3.2 Body and Head Manual Annotations
3.3 Feature Extraction
3.4 Classification
4 Experimental Results and Discussion
4.1 Experimental Setup
4.2 Datasets
4.3 Experiment Results
5 Conclusion and Future Directions
References
Comparative Analysis of CNN-Based Spatiotemporal Reasoning in Videos
1 Introduction
2 Related Work
3 Methodology
3.1 ST modeling Architecture
3.2 Multi-layer Perceptron (MLP) Based Techniques
3.3 Recurrent Neural Networks (RNN) Based Techniques
3.4 Fully Convolutional Network (FCN) Based Techniques
3.5 Training Details
4 Experiments
4.1 Datasets
4.2 Resource Efficiency Analysis
4.3 Results Using Jester Dataset
4.4 Results Using Something-Something Dataset
5 Conclusion
References
Generalization of Fitness Exercise Recognition from Doppler Measurements by Domain-Adaption and Few-Shot Learning
1 Introduction
2 Related Works
3 Ultrasonic Smartphone Exercises: Sensing and Database
3.1 Sensing Principle
3.2 Database
4 Methods
4.1 The Baseline Method
4.2 Our Proposed Method 1: Domain Adaptation
4.3 Our Proposed Method 2: Few-Shot Classification
5 Evaluation and Discussion
5.1 Baseline Results
5.2 Our Proposed Approaches
5.3 Discussion
6 Conclusion
References
Local Anomaly Detection in Videos Using Object-Centric Adversarial Learning
1 Introduction
2 Related Work
3 Proposed Method
3.1 Object Detection and Gradient Extraction
3.2 Gradient-Appearance Relation Discovery Network (GARDiN)
3.3 Partial Mean Squared Reconstruction Errors (PMSRE)
3.4 Adversarial Classification of the PMSRE
3.5 Abnormal Events Detection
4 Experiments
4.1 Datasets and Evaluation Procedure
4.2 Experimental Setup
4.3 Results
4.4 Ablation Study
4.5 Inference Running Time
5 Conclusion
References
A Hierarchical Framework for Motion Trajectory Forecasting Based on Modality Sampling
1 Introduction
2 Related Work
3 Methodology
3.1 Transformation Set Construction
3.2 Encoder
3.3 Decoder
4 Experimental Results
4.1 Multi-modality Evaluation on Synthetic Data
4.2 Trajectory Prediction Evaluation
5 Conclusions
References
Skeleton-Based Methods for Speaker Action Classification on Lecture Videos
1 Introduction
2 Background
2.1 Vision-Based Human Pose Estimator
2.2 Action Classification
2.3 Lecture Video Analysis
3 Methodology
3.1 Action Segment Extraction
3.2 Speaker Pose Estimation
3.3 Speaker Pose Normalization
3.4 Action Segment Representation
3.5 Speaker Action Classification
4 Datasets
4.1 AccessMath
4.2 Extended Dataset
5 Experiments
5.1 Pose Estimator Selection
5.2 Adaptations of the 2S-AGCN Model for Speaker Action Classification
5.3 Classifier: AM Pose Vs 2S-AGCN Model
6 Conclusion
References
IADS - Integrated Artificial Intelligence In Data Science
Workshop on Integrated Artificial Intelligence in Data Science (IADS)
Workshop Description
Organization
IADS Chairs
Program Committee
Fake Review Classification Using Supervised Machine Learning
1 Introduction and Background
1.1 Problem Statement
1.2 Research Questions
1.3 Aims and Objectives
1.4 Research Contributions
2 Literature Review
3 Methodology
3.1 Data Acquisition
3.2 Reducing Noise
3.3 Implementation of Supervised Machine Learning Classifier
3.4 Applying Other Machine Learning Classifiers
3.5 How the Proposed System Works?
3.6 Comparing Classifiers' Performance
4 Results
4.1 Answer to First Research Question
4.2 Answer to Second Research Question
4.3 Answer to Third Research Question
5 Conclusion and Future Work
References
Defect Detection of Stainless Steel Plates Using Deep Learning Technology
1 Introduction
2 Related Works
2.1 YOLOv3: You Only Look once V3
2.2 SSD: Single Shot MultiBox Detector
3 Stainless Steel Images and Definitions of Defects
3.1 Linear Defects
3.2 Regional Defects
4 Experiment
4.1 Image Capture and Environment Establishment
4.2 Dataset
4.3 YOLOv3 Model Training and SSD 300 Model Training
4.4 Verification of the Classification Accuracy of the YOLOv3 Model
4.5 Verification of the Classification Accuracy of the SSD 300 Model
5 Conclusion
References
Deep Neural Networks for Detecting Real Emotions Using Biofeedback and Voice
1 Introduction
1.1 Paper Goals and Organization
2 Related Works
3 Methodology
3.1 Performing the Regression
3.2 Building the Deep ANN
4 Experiment and Results
4.1 Techniques Used
4.2 Results
5 Conclusion
References
Data Augmentation for a Deep Learning Framework for Ventricular Septal Defect Ultrasound Image Classification
1 Introduction
2 Methods
2.1 Problem Definition
2.2 Data Augmentation Approaches
3 Empirical Results of Data Augmentation
3.1 Comparison of the Regression Models
3.2 Weight Values of Linear Regression Model
3.3 Comparisons of the Three DL Algorithms
4 Empirical Results of All VSD Type for a Larger Problem Size
5 Conclusions
References
A Neural Network Model for Lead Optimization of MMP12 Inhibitors
1 Introduction
2 Lead Optimization Problem
3 Lead Optimization Based on Predictive Neural Network Models
4 Results and Discussion
4.1 Evaluation Metrics
4.2 Dataset
4.3 Results of Lead Optimization Using SMPP
4.4 Results of Lead Optimization Using SPP
5 Conclusion and Future Directions
References
An Empirical Analysis of Integrating Feature Extraction to Automated Machine Learning Pipeline
1 Introduction
2 Reference Tools
2.1 Automated Feature Engineering Tool
2.2 AutoML Frameworks
3 Experimental Setup
3.1 Experiment Design
3.2 Datasets
4 Results and Discussion
5 Conclusion and Future Work
References
Input-Aware Neural Knowledge Tracing Machine
1 Introduction
2 Related Work
2.1 Factor Analysis
2.2 Knowledge Tracing Machine
2.3 DeepFM for Knowledge Tracing
3 Proposed Model
3.1 Encoding and Embedding Layer
3.2 Attention Net
3.3 Representation Refining and Sum Pooling Layer
3.4 Hidden Layers
3.5 Prediction Units
3.6 Optimization
4 Experiments
4.1 Datasets
4.2 Baselines
4.3 Student Performance Prediction
4.4 Ablation Study
5 Conclusion
References
Towards Corner Case Detection by Modeling the Uncertainty of Instance Segmentation Networks
1 Introduction
2 Related Work
3 Mask R-CNN
4 Bayesian Uncertainty Modeling
5 Our Approach for Corner Case Detection
5.1 Uncertainty Estimation
5.2 Corner Case Detection
6 Experimental Setup and Result Analysis
6.1 Dataset
6.2 Qualitative Evaluation
7 Conclusion and Future Improvements
References
Intelligent and Interactive Video Annotation for Instance Segmentation Using Siamese Neural Networks
1 Introduction
2 Related Work
3 Towards Intelligent Annotation
3.1 Base Model
3.2 Interaction Model
3.3 Weighted Binary Cross-Entropy Loss
3.4 Approaches to Incorporate Corrections
4 Evaluation Methodology
5 Evaluation
5.1 Weighted Binary Cross-Entropy Loss
5.2 Intersection over Union
6 Discussion
7 Conclusion and Future Work
References
Imputation of Rainfall Data Using Improved Neural Network Algorithm
1 Introduction
2 Study Region
3 Methodology
3.1 Phase 1: Data Preparation
3.2 Phase 2: Missing Data Imputation
4 Material and Methods
4.1 Meteorological Data
4.2 Rainfall from Nearest Neighbor Stations
4.3 Pre-processing Data Using Principal Component Analysis (PCA)
4.4 Imputation Approaches
4.5 Performance Measures
5 Results and Discussion
6 Conclusions
References
Novelty Based Driver Identification on RR Intervals from ECG Data
1 Introduction
2 Related Work
3 Data
4 Own Approach
4.1 Feature Extraction
4.2 Novelty Detection
4.3 Driver Identification
5 Experimental Evaluation
6 Conclusion and Outlook
References
Link Prediction in Social Networks by Variational Graph Autoencoder and Similarity-Based Methods: A Brief Comparative Analysis
1 Introduction
2 Link Prediction: Problem Description and Related Works
3 Methodology
3.1 Variational Graph Autoencoder
3.2 Similarity-Based Methods
4 Numerical Experiments
4.1 Data Set
4.2 Performance Metrics
4.3 Numerical Results
5 Conclusions
References
A Hybrid Wine Classification Model for Quality Prediction
1 Introduction
2 Background Knowledge
2.1 The SVM Classifier
2.2 Random Forest
3 Proposed Hybrid Wine Classification Model
4 Experimental Evaluation
4.1 Dataset Description
4.2 Performance Measure Metrics
4.3 Experimental Analysis
5 Conclusions and Future Work
References
A PSO-Based Sanitization Process with Multi-thresholds Model
1 Introduction
2 Related Work
3 Background Knowledge
4 Proposed Sanitization Algorithm
5 Experimental Results
5.1 Fitness Value
5.2 Runtime
6 Conclusions
References
Task-Specific Novel Object Characterization
1 Introduction
2 Related Work
3 Method
3.1 Detection of Common Objects
3.2 Task-Specific Taxonomy
3.3 Uncertainty and Relevancy for the Task
3.4 Characterizing Relevant Objects
4 Experiments
4.1 Demo Video on YouTube
4.2 Visualization
4.3 Finding the Relevant Objects
4.4 Ignoring the Irrelevant Objects
4.5 Summary of the Results
5 Conclusions
References
IML - International Workshop on Industrial Machine Learning
Workshop on Industrial Machine Learning (IML)
Workshop Description
Organization
IML Chairs
Program Committee
Deep Learning Based Dimple Segmentation for Quantitative Fractography
1 Introduction
2 Background and Related Work
2.1 Computer Vision in Quantitative Fractography
2.2 Fractography
2.3 Crack Growth
2.4 Microstructure
2.5 Dimple Fracture
3 Methodology
3.1 Previous Approaches
3.2 Proposed Model
4 Evaluation
4.1 Dataset
4.2 Experiments
4.3 Results
5 Conclusion
References
PaDiM: A Patch Distribution Modeling Framework for Anomaly Detection and Localization
1 Introduction
2 Related Work
3 Patch Distribution Modeling
3.1 Embedding Extraction
3.2 Learning of the Normality
3.3 Inference: Computation of the Anomaly Map
4 Experiments
4.1 Datasets and Metrics
4.2 Experimental Setups
5 Results
5.1 Ablative Studies
5.2 Comparison with the State-of-the-art
5.3 Anomaly Localization on a Non-aligned Dataset
5.4 Scalability Gain
6 Conclusion
References
Real-Time Cross-Dataset Quality Production Assessment in Industrial Laser Cutting Machines
1 Introduction
2 Problem Description
3 Related Works
4 Inference Model
4.1 Architectures
4.2 Adaptation
5 Deployment
5.1 Client Server Communication
5.2 Hardware
6 Experiments
6.1 Datasets
6.2 Training Procedure
6.3 Results
7 Discussion
8 Conclusion
References
An Online Deep Learning Based System for Defects Detection in Glass Panels
1 Introduction
2 Related Works
3 Problem Description
3.1 Deltamax Detection System
3.2 CLEVER Glass Dataset
4 Methodology
4.1 Overall System Configuration
4.2 Preprocessing
5 Results
5.1 Architecture Selection
5.2 Channel Shift Analysis
5.3 Discussion
5.4 Qualitative Results: Single-Channel, Multi-channel
6 Conclusion
References
Evaluation of Edge Platforms for Deep Learning in Computer Vision
1 Introduction
2 Related Work
2.1 Object Classification
2.2 Object Detection
2.3 Semantic Segmentation
2.4 Platform Benchmarks
3 Platform Evaluation
3.1 Model Overview
3.2 Platform Overview
3.3 Evaluation Overview
4 Experimental Results
4.1 Classification
4.2 Object Detection
4.3 Semantic Segmentation
4.4 Comparison of Tasks
4.5 Inference Analysis
5 Conclusion
References
BlendTorch: A Real-Time, Adaptive Domain Randomization Library
1 Introduction
1.1 Related Work
1.2 Contributions
2 Design Principles
3 Architecture
4 Experiments
4.1 Datasets
4.2 BlendTorch Dataset
4.3 Training
4.4 Prediction
4.5 Average Precision
4.6 Runtime
5 Conclusion
References
SAFFIRE: System for Autonomous Feature Filtering and Intelligent ROI Estimation
1 Introduction
2 Related Work
3 Materials and Methods
3.1 Training Data and Reference Image
3.2 Feature Extraction
3.3 The Training Graph
3.4 Search of Optimal Path in the Training Graph
3.5 The Anchor
3.6 Selection of the Best Feature Type
3.7 Testing
4 Evaluation Setup
4.1 Evaluation Datasets
4.2 Evaluation Metrics
5 Results
5.1 Noise Patterns and Filtering
5.2 Location Accuracy
5.3 Computation Time
6 Discussions
7 Conclusions
References
Heterogeneous Feature Fusion Based Machine Learning on Shallow-Wide and Heterogeneous-Sparse Industrial Datasets
1 Introduction
2 Background
2.1 Transfer Learning
2.2 Heterogeneous Feature Fusion
3 Method
3.1 Proposed Method
3.2 Baseline Methods
3.3 Evaluation Method
4 Results and Discussion
4.1 Toyota Dataset No. 1
4.2 Toyota Dataset No. 2
4.3 Discussion
5 Conclusion
References
3-D Deep Learning-Based Item Classification for Belt Conveyors Targeting Packaging and Logistics
1 Introduction
2 Background
2.1 Conveyor-Based Item Classification System
2.2 Machine Learning
2.3 Deep Learning
2.4 3-D Deep Learning Models
3 Experimental Setup
3.1 Hardware Setup
3.2 Software Setup
3.3 Items Used
4 Methods
4.1 Dataset Acquisition from Sensor Controller
4.2 Point Cloud Equalization
4.3 Dataset Generation
4.4 Candidate Classification Models
4.5 Model Training Strategy
5 Experimental Results
6 Discussion and Conclusion
References
Development of Fast Refinement Detectors on AI Edge Platforms
1 Introduction
2 Baseline Architecture - RefineDet
3 Combining Light-Weight Backbone CNNs
4 Evaluation Environments
5 Results and Discussions
6 Conclusions and Future Works
References
Selecting Algorithms Without Meta-features
1 Introduction
2 Background
2.1 Semantic Segmentation and Object Detection
2.2 Algorithm Selection (AS)
3 Data Set for Algorithm Selection
3.1 Ideal Algorithm Selection
4 Experiments
4.1 Experiments on VOC2012
4.2 Experiments on MSCOCO
4.3 Dynamic Ensemble Approach
5 Results and Discussion
6 Conclusion
References
A Hybrid Machine Learning Approach for Energy Consumption Prediction in Additive Manufacturing
1 Introduction
2 Literature Review
2.1 Analysis of Energy Consumption in AM Systems
2.2 Machine Learning for Predictive Modelling in AM System
3 Methodology
3.1 Multi-source Data Sensing and Collection
3.2 The Hybrid ML Approach for Energy Consumption Prediction
3.3 Validation of Prediction Model
4 Case Study
4.1 Experimental Setup
4.2 Results and Discussion
5 Conclusions
References
Bias from the Wild Industry 4.0: Are We Really Classifying the Quality or Shotgun Series?
1 Introduction
2 Related Work
3 Dataset
4 Task Definition
5 Classification Model
5.1 CNN Architectures
5.2 Loss Functions
6 Experimental Procedure
6.1 Training Settings
7 Results
7.1 Classification Performance
7.2 Bias Detection
7.3 Bias Mitigation
8 Conclusions
References
Machine Learning for Storage Location Prediction in Industrial High Bay Warehouses
1 Introduction
2 Related Work and Background
3 Methods and Results
3.1 Logistic Data Model
3.2 Retrieving ABC Classes for Training Data
3.3 Extending Feature Set
3.4 Implementation / Decision Tree
4 Evaluation and Discussion
5 Conclusion
References
A Deep Learning-Based Approach for Automatic Leather Classification in Industry 4.0
1 Introduction
2 Related Works
3 Materials and Methods
3.1 LASCC Dataset
3.2 Deep Learning Models
3.3 Performance Evaluation
4 Results and Discussions
5 Conclusions and Future Works
References
Automatic Viewpoint Estimation for Inspection Planning Purposes
1 Introduction
2 Related Work
2.1 Inspection Viewpoint Selection
2.2 Pose from RGB/RGB-D Data
3 Approach
3.1 Data Generation
3.2 Architecture
3.3 Training Process
3.4 Runtime
4 Experiments and Results
4.1 Experimental Setup
4.2 Testing Using Different Objects
4.3 Grayscale Images
4.4 Effect of Size of Bottleneck Layer
4.5 Effect of Cluster Numbers on the Accuracy and Performance
4.6 Reconstructed Images
4.7 Inference Time
4.8 Failure Cases
5 Conclusion
References
Localisation of Defects in Volumetric Computed Tomography Scans of Valuable Wood Logs
1 Introduction
2 Our Approach
2.1 2D Defect Detection
2.2 Multi-defect Tracking
3 Experimental Results
3.1 Data Acquisition and Normalization
3.2 Datasets
3.3 Experimental Setup
3.4 Analysis of the Results
4 Conclusions
References
Image Anomaly Detection by Aggregating Deep Pyramidal Representations
1 Introduction
2 Related Work
3 Proposed Model
3.1 Network Architecture
3.2 Objective and Losses
4 Experimental Results
5 Ablation Study
6 Conclusions
References
Fault Detection in Uni-Directional Tape Production Using Image Processing
1 Introduction
2 Related Work
3 Proposed Method
3.1 Tape Location
3.2 Void Detection
3.3 Void Classification
4 Evaluation and Results
4.1 Tape Location Results
4.2 Void Detection Results
4.3 Void Classification Results
5 Conclusion
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken