Graph Learning for Fashion Compatibility Modeling

Description

This book sheds light on state-of-the-art theories for more challenging outfit compatibility modeling scenarios.  In particular, this book presents several cutting-edge graph learning techniques that can be used for outfit compatibility modeling.  Due to its remarkable economic value, fashion compatibility modeling has gained increasing research attention in recent years.  Although great efforts have been dedicated to this research area, previous studies mainly focused on fashion compatibility modeling for outfits that only involved two items and overlooked the fact that each outfit may be composed of a variable number of items.  This book develops a series of graph-learning based outfit compatibility modeling schemes, all of which have been proven to be effective over several public real-world datasets.  This systematic approach benefits readers by introducing the techniques for compatibility modeling of outfits that involve a variable number of composing items.  To deal with the challenging task of outfit compatibility modeling, this book provides comprehensive solutions, including correlation-oriented graph learning, modality-oriented graph learning, unsupervised disentangled graph learning, partially supervised disentangled graph learning, and metapath-guided heterogeneous graph learning.  Moreover, this book sheds light on research frontiers that can inspire future research directions for scientists and researchers.   Preface Acknowledgements Contents About the Authors 1 Introduction 1.1 Background 1.2 Challenges 1.3 Our Solutions 1.4 Book Structure 2 Correlation-Oriented Graph Learning for OCM 2.1 Introduction 2.2 Related Work 2.3 Methodology 2.3.1 Problem Formulation 2.3.2 Multimodal Outfit Compatibility Modeling 2.4 Experiment 2.4.1 Experimental Settings 2.4.2 Model Comparison 2.4.3 Ablation Study 2.4.4 Case Study 2.5 Summary 3 Modality-Oriented Graph Learning for OCM 3.1 Introduction 3.2 Related Work 3.3 Methodology 3.3.1 Problem Formulation 3.3.2 Multimodal Embedding 3.3.3 Modality-Oriented Graph Learning 3.3.4 Outfit Compatibility Estimation 3.4 Experiment 3.4.1 Experimental Settings 3.4.2 Model Comparison 3.4.3 Ablation Study 3.4.4 Modality Comparison 3.4.5 Hyperparameter Discussion 3.5 Summary 4 Unsupervised Disentangled Graph Learning for OCM 4.1 Introduction 4.2 Related Work 4.3 Methodology 4.3.1 Problem Formulation 4.3.2 Context-Aware Outfit Representation Learning 4.3.3 Hidden Complementary Factors Learning 4.3.4 Outfit Compatibility Modeling 4.4 Experiment 4.4.1 Experimental Settings 4.4.2 Model Comparison 4.4.3 Ablation Study 4.4.4 Case Study 4.5 Summary 5 Supervised Disentangled Graph Learning for OCM 5.1 Introduction 5.2 Related Work 5.3 Methodology 5.3.1 Problem Formulation 5.3.2 Partially Supervised Compatibility Modeling 5.4 Experiment 5.4.1 Experimental Settings 5.4.2 Model Comparison 5.4.3 Ablation Study 5.4.4 Case Study 5.5 Summary 6 Heterogeneous Graph Learning for Personalized OCM 6.1 Introduction 6.2 Related Work 6.3 Methodology 6.3.1 Problem Formulation 6.3.2 Metapath-Guided Personalized Compatibility Modeling 6.4 Experiment 6.4.1 Experimental Settings 6.4.2 Model Comparison 6.4.3 Ablation Study 6.4.4 Sensitivity Analysis 6.4.5 Case Study 6.5 Summary 7 Research Frontiers [DELETE] 7.1 Efficient Fashion Compatibility Modeling 7.2 Unbiased Fashion Compatibility Modeling 7.3 Try-On Enhanced Fashion Compatibility Modeling