Computer Animation and Social Agents: 33rd International Conference on Computer Animation and Social Agents, CASA 2020, Bournemouth, UK, October 13-15, 2020, Proceedings

Preface
Organization
Contents
Modelling, Animation and Simulation
ESENet: A Human Behavior Recognition Model Based on Extended Squeeze-and-Excitation Network
1 Introduction
2 Related Work
3 Methods
3.1 ESENet
3.2 The Integration of 3DSE and 3DCNN
4 Experiments and Analysis
4.1 The Details of Datasets and Our Experiment
4.2 The Ablation Study
4.3 Comparing 3DSE with SE
4.4 The Comparison to Other Approaches on Different Datasets
5 Conclusion
References
Interpolating Frames for Super-Resolution Smoke Simulation with GANs
1 Introduction
2 Related Work
3 Proposed Approach
3.1 Interpolating Frames
3.2 Slow Fusion
3.3 Loss Functions
4 Data Generation and Training
5 Results
6 Conclusions and Future Work
References
3D Face Reconstruction and Semantic Annotation from Single Depth Image
1 Introduction
2 Method
2.1 Coarse-Level Facial Code Inference
2.2 Fine-Level Displacement Inference
2.3 Semantic Annotation
2.4 Loss
3 Experiments
4 Conclusion
References
A Robot for Interactive Glove Puppetry Performance
1 Introduction
2 Related Work
3 Method
3.1 Robot
3.2 Interactive Performance
4 Experience
4.1 Performance Configuration
4.2 Interaction Sections in the Performance
4.3 Analysis
5 Conclusion
References
Virtual Reality
Efficient Metaballs-Based Collision Detection for VR Neurosurgery Simulation on GPU
1 Introduction
2 Related Work
3 Construction and Optimization of the Inner Metaballs
3.1 Initialization of the Inner Metaballs
3.2 Global Optimization Using the Electrostatic Attraction Model
4 Collision Detection and Response
4.1 Self and Multi-Organs' Collision
4.2 Adaptive Cylinder Based Collision Detection
4.3 Collision Response and Deformation
5 Experiments, Comparison, and Application
6 Conclusion and Discussion
References
Posing 3D Characters in Virtual Reality Through In-the-Air Sketches
1 Introduction
2 Related Works
3 Proposed System
3.1 VR-Based Environment for 3D Sketching
3.2 Matching and Aligning Algorithm
3.3 Current Development State
4 Conclusions and Future Work
References
Evaluating the Need and Effect of an Audience in a Virtual Reality Presentation Training Tool
1 Introduction
2 Related Work
2.1 VR as a Presentation Environment
3 The VR Presentation Environment
4 Experiment Design
4.1 Environment Metrics
4.2 Apparatus
4.3 Participants
4.4 Procedures
5 Results
5.1 Pre-questionnaire
5.2 Questionnaires of Nervousness and Preference Order
5.3 Open-Ended Questions
5.4 Discussion
6 Conclusion
References
Image Processing and Computer Vision
A Cascaded Approach for Keyframes Extraction from Videos
1 Introduction
2 Related Work
3 Method
3.1 Overview
3.2 Generating Initial Clips
3.3 Adaptive Window Length
3.4 Meaningless Frames Removal
3.5 Mapping and Optimization
4 Experiments
4.1 Experimental Setup
4.2 Quantitative Results
5 Conclusion
References
Extracting Highlights from a Badminton Video Combine Transfer Learning with Players' Velocity
1 Introduction
2 Related Work
3 The Proposed Method
3.1 Badminton Video Segmentation
3.2 Highlight Extraction
4 Experiment
4.1 Badminton Video Segmentation
4.2 Evaluation of Extracted Highlights
5 Conclusions
References
Dynamic Human Body Size Measurement Based on Feature Points Prediction and Mapping
1 Method
1.1 Random Forest Regression Analysis
1.2 3D Human Body Mapping
1.3 Measurement of Models
2 Experiment
2.1 Parameter Setting of Random Forest Regression
2.2 Correspondence Map
2.3 Measurement of 3D Human Models
3 Conclusion
References
Amur Tiger Re-ID Based on Locally Guided Adaptive Feature Fusion
1 Introduction
2 Related Work
3 Methodology
3.1 Partial Image
3.2 Locally Guided Adaptive Feature Fusion Network
3.3 Feature Fusion
3.4 Loss Function
4 Experiment
4.1 Dataset
4.2 Experiment Details
4.3 Experimental Results
5 Conclusion
References
MVFNN: Multi-Vision Fusion Neural Network for Fake News Picture Detection
1 Introduction
2 Related Works
3 The Proposed Model
3.1 Model Overview
3.2 Model Derivation
4 Experiments
4.1 Datasets
4.2 Baselines
4.3 Performance Comparison
4.4 Ablation Study
4.5 Multi-feature Fusion
5 Conclusion
References
PEFS: A Validated Dataset for Perceptual Experiments on Face Swap Portrait Videos
1 Introduction
2 Related Work
3 Dataset
4 Perceptual Validation
5 Conclusion
References
Altering the Conveyed Facial Emotion Through Automatic Reenactment of Video Portraits
1 Introduction and Related Work
2 Reenacting Technique
3 Experimental Design
4 Results and Discussion
5 Conclusion
References
Author Index