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.
1 Introduction1.1 Background and Motivation1.2 Overview1.3 Acronyms and Notations1.4 Roadmap2 Literature Review2.1 Event Understanding2.2 Tag Recommendation and Ranking2.3 Soundtrack Recommendation for UGVs2.4 Lecture Video Segmentation3 Event Understanding3.1 Introduction3.2 System Overview3.2.1 EventBuilder3.2.2 EventSensor3.3 Evaluation3.3.1 EventBuilder3.3.2 EventSensor3.4 Summary4 Tag Recommendation and Ranking4.1 Introduction4.1.1 Tag Recommendation4.1.2 Tag Ranking4.2 System Overview4.2.1 Tag Recommendation4.2.2 Random Walk based Relevance Scores4.2.3 Fusion of Different Tag Recommendation Approaches4.2.4 Tag Ranking4.3 Evaluation4.3.1 Tag Recommendation4.3.2 Tag Ranking4.4 Summary5 Soundtrack Recommendation for UGVs5.1 Introduction5.1.1 Increasing Popularity of User-Generated Videos5.1.2 Challenges with User-Generated Videos in Viewing and Sharing5.1.3 Motivation for Generating Music Videos for Outdoor
User-Generated Videos5.2 Music Video Generation5.2.1 Scene Moods Prediction Models5.2.2 Music Retrieval Techniques5.2.3 Automatic Music Video Generation Model5.3 Evaluation5.3.1 Dataset and Experimental Settings5.3.2 Evaluation Metrics5.3.3 Objective Evaluation5.3.4 Subjective Evaluation5.3.5 Experimental Results5.3.6 Comparison with State-of-the-arts5.3.7 Discussion of Results5.4 Summary6 Lecture Video Segmentation6.1 Introduction6.2 Lecture Video Segmentation6.2.1 Prediction of Video Transition Cues using Supervised Learning6.2.2 Computation of Text Transition Cues using N-gram based Language Model6.2.3 Computation of SRT Segment Boundaries using the state-of-the-art6.2.4 Computation of Wikipedia Segment Boundaries6.2.5 Transition File Generation<6.3 Evaluation6.3.1 Dataset and Experimental Settings6.3.2 Results from the ATLAS System6.3.3 Results from the TRACE System6.4 Summary7 Conclusions and future work
Introduction.- Literature Review.- Event Understanding.- Tag Recommendation and Ranking.- Soundtrack Recommendation for UGVs.- Lecture Video Segmentation.- Adaptive News Video Uploading.- Conclusions and future work.
Rajiv Ratn Shah received his B.Sc. with honors in Mathematics from Banaras Hindu University, India in 2005. He received his M.Tech. in Computer Technology and Applications from Delhi Technological University, India in 2010. Prior joining Indraprastha Institute of Information Technology Delhi (IIIT Delhi), India as an assistant professor, Dr Shah has received his Ph.D. in Computer Science from the National University of Singapore, Singapore. Currently, he is also working as a research fellow in living analytics research centre (LARC) at the Singapore Management University, Singapore. His research interests include the multimodal analysis of user-generated multimedia content in the support of social media applications, multimodal event detection and recommendation, and multimedia analysis, search, and retrieval. Dr Shah is the recipient of several awards, including the runner-up in the Grand Challenge competition of ACM International Conference on Multimedia. He is involved in reviewingof many top-tier international conferences and journals. He has published several research work in top-tier conferences and journals such as Springer MultiMedia Modeling, ACM International Conference on Multimedia, IEEE International Symposium on Multimedia, and Elsevier Knowledge-Based Systems.
Daugiau apie elektronines knygas