Multimodal Analysis of User-Generated Multimedia Content (Socio-Affective Computing)

This book presents a summary of the multimodal analysis of user-generated multimedia content (UGC). Several multimedia systems and their proposed frameworks are also discussed. First, improved tag recommendation and ranking systems for social media photos, leveraging both content and contextual information, are presented. Next, we discuss the challenges in determining semantics and sentics information from UGC to obtain multimedia summaries. Subsequently, we present a personalized music video generation system for outdoor user-generated videos. Finally, we discuss approaches for multimodal lecture video segmentation techniques. This book also explores the extension of these multimedia system with the use of heterogeneous continuous streams.