Modeling Visual Aesthetics, Emotion, and Artistic Style

• Autore: Wang James Z
• Pubblicazione: Cham : , : Springer International Publishing AG, , 2024; ©2024
• Edizione: 1st ed.
• Descrizione fisica: 1 online resource (408 pages)
• Altri autori: AdamsReginald B., Jr
• Nota di contenuto: Intro -- Foreword -- Preface -- Acknowledgments -- Contents -- Contributors -- Acronyms -- Part I Foundations of Emotion Modeling and Machine Learning -- 1 Models of Human Emotion and Artificial Emotional Intelligence -- 1.1 Introduction -- 1.2 Emotion Modeling: A Psychological Perspective -- 1.2.1 Distinguishing Emotion from Other Affective Phenomena -- 1.2.2 Three Competing Theories -- 1.2.2.1 Basic Emotion Theory -- 1.2.2.2 Continuous Dimensions of Emotion -- 1.2.2.3 Componential Theories of Emotion -- 1.2.3 Cultural Considerations -- 1.2.4 Physiological Considerations -- 1.3 Applications in Computing -- 1.3.1 Emotion as a Target for Training -- 1.3.2 Mimicking Human Emotion -- 1.3.3 Statistical Techniques for Developing Emotion Models -- 1.3.4 Emergent AEI -- 1.4 Conclusion -- References -- 2 A Concise Introduction to Machine Learning -- 2.1 Learning Algorithms -- 2.1.1 The Task, T -- 2.1.2 The Performance Measure, P -- 2.1.3 The Experience, E -- 2.2 Evaluation and Model Selection -- 2.2.1 Overfitting, Underfitting, and Model Capacity -- 2.2.2 Bias and Variance -- 2.2.3 The No Free Lunch Theorem -- 2.2.4 Regularization -- 2.2.5 Parameter Tuning and Validation -- 2.3 Supervised Learning Algorithms -- 2.3.1 Linear Models -- 2.3.2 Support Vector Machine -- 2.4 Neural Networks and Deep Learning -- 2.4.1 Feedforward Networks -- 2.4.1.1 Perceptron -- 2.4.1.2 Multi-layer Perceptron -- 2.4.2 Back Propagation of Errors -- 2.4.3 Architecture Design of Neural Networks -- 2.5 Concluding Remarks -- References -- Part II Human Social Vision -- 3 Facing a Perceptual Crossroads: Mixed Messages and Shared Meanings in Social Visual Perception -- 3.1 Introduction -- 3.2 The Role of Eye Gaze in Emotion Perception -- 3.3 Gender/Sex and Emotion -- 3.4 Race and Emotion -- 3.5 Other Intersections -- 3.5.1 Who Else Is Missing?. ; 3.5.2 Other Identities to Consider? Future Directions -- 3.6 Conclusions -- References -- 4 Social Vision of the Body in Motion: Interactions Between the Perceiver and the Perceived -- 4.1 Introduction -- 4.2 Determinants of Body Perception Originating in the Target of Perception Dynamic Cues -- 4.2.1 Dynamic Cues -- 4.2.2 Structural Cues -- 4.3 Integration of Visible Cues -- 4.4 Perceiver Influences -- 4.4.1 Self-protective Biases -- 4.4.2 Knowledge Structures and Stereotypes -- 4.5 Deliberate Manipulation of Own Body Movement -- 4.6 Conclusion -- References -- 5 Visual Perception of Threat: Structure, Dynamics, and Individual Differences -- 5.1 Introduction -- 5.2 How Do We Distinguish Threat from Other Negative Stimuli? -- 5.3 Threat Dimensions -- 5.3.1 Threat Direction -- 5.3.2 Threat Imminence -- 5.3.3 Threat Capacity -- 5.3.4 Evidence for the Three Threat Dimensions -- 5.4 The Role of Context in Threat Perception -- 5.5 Hemispheric Lateralization in Threat Perception -- 5.6 Clear and Ambiguous Threat in Faces -- 5.7 The Role of the Major Visual Pathways in Threat Perception -- 5.8 Individual Differences -- 5.8.1 Threat and Anxiety -- 5.8.2 Sex Differences in Threat Processing -- 5.9 Summary -- References -- 6 From Pixels to Power: Critical Feminist Questions for the Ethics of Computer Vision -- 6.1 Introduction -- 6.2 Feminist Approaches -- 6.2.1 Intersectional Feminisms -- 6.2.2 Standpoint Feminisms -- 6.3 Algorithmic Bias -- 6.4 Invisible Labor -- 6.5 Resistance -- 6.5.1 Responses Within -- 6.5.2 Activist Responses -- 6.6 Conclusion -- References -- Part III Computer Social Vision -- 7 High-Speed Joint Learning of Action Units and Facial Expressions -- 7.1 Introduction -- 7.2 Related Work -- 7.3 The Algorithm -- 7.3.1 Problem Formulation -- 7.3.2 Optimization -- 7.3.3 Joint Learning of Action Units and Facial Expressions. ; 7.3.4 Efficiency Advantage -- 7.4 Experiments -- 7.4.1 Experimental Setup -- 7.4.2 Performance Comparison -- 7.4.2.1 AU Recognition -- 7.4.2.2 Emotion Recognition -- 7.4.3 Effect of Model Choice -- 7.4.4 Computational Speed -- 7.4.5 Tagging a Real-World Dataset -- 7.5 Conclusion and Future Work -- References -- 8 ExpressionFlow: A Microexpression Descriptor for Efficient Recognition -- 8.1 Introduction -- 8.2 Related Work -- 8.3 Our Method -- 8.3.1 The ExpressionFlow -- 8.3.2 The Optimization Process -- 8.3.3 The Algorithm -- 8.3.4 Comparison to Facial Dynamics Map -- 8.3.5 Computational Complexity -- 8.4 Experiments -- 8.4.1 Experiments Setup -- 8.4.2 Recognition Results -- 8.4.3 Effects of Parameters -- 8.4.4 Runtime -- 8.5 Conclusions and Future Work -- References -- 9 Emotion in the Neutral Face: Applications for Computer Vision and Aesthetics -- 9.1 Introduction -- 9.2 Definitions of ``Neutral'' -- 9.3 ``Neutral'' in Person Perception -- 9.3.1 Neutral is Disturbing -- 9.3.2 True Neutral Is Impossible -- 9.4 ``Neutral'' in Computer Vision -- 9.5 Bias in Computer Vision -- 9.6 Broader Implications -- 9.7 Conclusions -- References -- 10 Multi-Stream Temporal Networks for Emotion Recognition in Children and in the Wild -- 10.1 Introduction -- 10.2 Related Work -- 10.3 Multi-Stream Architectures -- 10.3.1 Temporal Segment Networks -- 10.3.2 First Multi-Stream Architecture-Emotion Recognition in the Wild -- 10.3.2.1 Body and Context -- 10.3.2.2 Embedding Loss -- 10.3.2.3 Predictions -- 10.3.3 Second Multi Stream Architecture-Child Emotion Recognition -- 10.3.3.1 Face -- 10.3.3.2 Audio Branch -- 10.3.3.3 Training and Audiovisual Fusion -- 10.4 Experimental Results -- 10.4.1 First Use-case: Multi-Stream Emotion Recognition In-the-wild -- 10.4.2 Second Use-Case: Child Emotion Recognition -- 10.4.2.1 Number of Segments. ; 10.4.2.2 Audiovisual Fusion and Training Schemes -- 10.4.2.3 Emotion by Modality -- 10.4.2.4 Final Results -- 10.5 Conclusions and Future Work -- References -- Part IV Photography, Arts -- 11 The Formal Language of Photography: A Primer -- 11.1 Introduction -- 11.2 Primary Visual Elements -- 11.2.1 Tone and Color -- 11.2.2 Tonality -- 11.2.3 Tone Contrast -- 11.2.4 Color -- 11.2.5 Color Contrast -- 11.3 Secondary Visual Elements -- 11.3.1 Lines -- 11.3.2 Shapes -- 11.3.3 Textures -- 11.4 Representing Three-Dimensional Space -- 11.4.1 Light -- 11.4.2 Depth and Dimension -- 11.4.3 Volume -- 11.5 Photography-Specific Elements -- 11.5.1 Vantage Point -- 11.5.2 Frame -- 11.5.3 Time and Movement -- 11.5.4 Focus and Depth of Field -- 11.6 Composition -- 11.6.1 Harmony and Contrast -- 11.6.2 Balance and Instability -- 11.6.3 Repetition, Irregularity, Unity, Variety -- 11.6.4 Simplicity, Complexity, Emphasis -- 11.7 Beyond Form -- 11.7.1 Form and Contents -- 11.7.2 Contents and Subject -- 11.7.3 The Photograph as Object -- References -- 12 Breathing with Robots: Notating Performer Strategy, Alongside Choreographer Intent and Audience Observation, in Breath-Driven Robotic Dance Performance -- 12.1 Introduction -- 12.2 Breath -- 12.3 Breath Patterns in Babyface -- 12.4 Embodied Exercises for Deeper Understanding -- 12.5 Choreographer, Performer, and Audience Analysis -- 12.6 Relating Breath to Robots in Public Spaces -- 12.7 Conclusion -- References -- 13 Humanist-in-the-Loop: Machine Learning and the Analysis of Style in the Visual Arts -- 13.1 Introduction -- 13.2 Debating Style -- 13.3 Humanist-in-the-Loop -- References -- Part V Aesthetics -- 14 The Inter-Relationship Between Photographic Aesthetics and Technical Quality -- 14.1 Introduction -- 14.1.1 Overview -- 14.1.2 Background -- 14.2 Interconnections in Representations. ; 14.2.1 Traditional Feature Representations -- 14.2.2 Deep Feature Representations -- 14.3 Prediction -- 14.3.1 Within- and Cross-Domain Testing -- 14.3.2 Feature Importance -- 14.3.3 Feature Selection -- 14.4 Discussion -- 14.4.1 On Factors Involved in Quality and Aesthetics Judgments -- 14.4.2 Psychometrics to the Rescue -- 14.5 Conclusions -- References -- 15 Image Restoration for Beautification -- 15.1 Introduction -- 15.2 Problem Definition -- 15.3 Method Design -- 15.3.1 Model Framework -- 15.3.1.1 Common Pipeline -- 15.3.1.2 Special Blocks -- 15.3.1.3 Generative Prior -- 15.3.2 Learning Strategies -- 15.3.2.1 Loss Function -- 15.3.2.2 Contrastive Learning -- 15.3.2.3 Semi-supervised Learning and Self-supervised Learning -- 15.3.2.4 Data Augmentation -- 15.4 Performance Evaluation -- 15.4.1 Dataset -- 15.4.1.1 Synthetic Data -- 15.4.1.2 Real Data -- 15.4.2 Evaluation Metric -- 15.5 Conclusions and Future Directions -- References -- 16 Image Affect Modeling: An Industrial Perspective -- 16.1 Overview -- 16.2 Observations of Image Affect Modeling -- 16.3 Use Cases of Image Affect Modeling -- 16.4 Opportunities of the Use of Image Affect Modeling -- 16.5 Conclusions -- References -- Part VI Emotion -- 17 Emotional Expression as a Means of Communicating Virtual Human Personalities -- 17.1 Introduction -- 17.2 Background -- 17.3 Related Work -- 17.3.1 Emotion and Personality Recognition -- 17.3.2 Emotion and Personality Synthesis -- 17.4 The Effect of Emotions on Personality Perception -- 17.4.1 Study 1: Emotional Facial Expressions and Personality Perception -- 17.4.2 Study 2: Emotional Body Poses and Personality Perception -- 17.5 Discussion -- 17.6 Conclusion -- References -- 18 Modeling Emotion Perception from Body Movements for Human-Machine Interactions Using Laban Movement Analysis -- 18.1 Introduction. ; 18.2 Emotion and Its Relation to the Body.
• Titolo autorizzato: Modeling Visual Aesthetics, Emotion, and Artistic Style
• ISBN: 3-031-50269-8
• Formato: Materiale a stampa
• Livello bibliografico: Monografia
• Lingua di pubblicazione: Inglese
• Record Nr.: 9910847082503321
• Lo trovi qui: Univ. Federico II