Blog‎ > ‎

Social and Mobile Media Workshop

posted Jan 30, 2013, 7:33 AM by Christian Timmerer
at IEEE International Conference on Multimedia and Expo
San Jose, USA. July 15 - 19, 2013


Multimedia tools and applications are flourishing faster than ever, as mobile and social media platforms became ubiquitous in the current decade. The rise of tablets and smart phones has enabled users across all age to easily create, share, and consume multimedia contents. At the same time, people are increasingly dependent on social and cloud services for obtaining information while they are on the move. While it seems like a straightforward correlation of supply and demand to sustain the rapidly growing multimedia data, there are many challenges and limitations that prohibit the full realization of its potentials. Mobility and social requirements for emerging applications, such as health, transport, intelligent environment, and adaptive community, drive the need to innovate the solutions for semantic extraction, indexing, summarization, and delivery of context-aware services. Social Multimedia refers to multimedia content generated from social networks and is considerably rich in meta-data information, including location of content capture, camera properties, user profile information from the social network, and textual descriptors (e.g., hash tags on Twitter). Such meta-data provides a natural context to multimedia, shrinking the semantic gap between the information that one can extract from visual data and the interpretation that the same data holds for a user; thus enabling improved content analysis.

A Social Multimedia Signal presumes human users as sensors and contains the spatio-temporal activity pattern of users with respect to some multimedia content. However, human interactions are known to be extremely diverse and often inconsistent, tailored at deep cognitive levels. Therefore, social multimedia data is inherently noisy - every social interaction/activity cannot be deemed worthy of bearing adequate information to contextualize the related media. Sampling, estimation and analysis of social multimedia signals could provide valuable information about spread of shared multimedia, communities that have affinity to some category of multimedia and temporal evolution of the social network.

This workshop aims to congregate researchers, developers and business (advertising/marketing) individuals to share their experiences, achievements and vision in exploring the potential of social multimedia signals collected from social network data, augmenting traditional multimedia and inventing novel frameworks of media collaboration through social signals.

The workshop invites original technical papers that were not previously published and are not currently under review for publication elsewhere. Topics include, but are not limited to:
  • Hot topics in the opportunities and challenges of mobile socio-media applications/services, from user and/or technology perspectives
  • Effective and streamlined retrieval models, query processing, and search interfaces for large-scale mobile multimedia applications
  • Cloud support for large scale socio-mobile media analysis Summarisation, content representation, and visualisation of multimedia contents for pervasive devices.a
  • Emerging architecture and systems to enable natural, seamless, smooth, and reliable multimedia delivery over wireless networks, including next generation streaming capabilities and new devices for inputs/outputs.
  • Interaction design and user modelling for multimedia usage in mobility and social settings
  • Innovative applications in mobile social multimedia, including the design and evaluation of the systems. For example, user-centered quality delivery of streaming mobile videos
  • Social Multimedia Signal Detection: what social activity traces can be considered useful signals for providing context to multimedia.
  • Social Multimedia Signal Sampling: theoretical and empirical studies of optimal sampling strategies for social multimedia signals
  • Social Multimedia Signal Estimation: quantifying an underlying phenomenon by measuring the temporal variation of the signal strength. For example, variation of trends signal in NY might provide information about the attention shifts/volatility of user groups in NYe
  • Cross-Domain Social Multimedia Signal Penetration: Does one type of social multimedia signal have a causal impact on another type when they exist in cross domains on the Web?
  • Social Multimedia Signal Encoding: Are any key properties of the social network graph encoded in the social multimedia signal?
  • Network Evolution Analysis using Social Multimedia Signals: Studying the temporal social multimedia signals might provide us with information about how the underlying social network is evolving with time.
  • Applications of social multimedia signals in multimedia information retrieval, communications, interactive visualization and multimedia modeling.
Important Dates
  • Full paper (6 pages) submission: March 7, 2013 
  • Notification of acceptance: April 15, 2013 
  • Camera ready papers due: April 30, 2013 
  • Workshop date: July 15 or 19, 2013 
Workshop Co-Chairs
  • Suman Deb Roy, University of Missouri, USA 
  • Dian Tjondronegoro, Queensland University of Technology, Australia 
  • Wenjun (Kevin) Zeng, University of Missouri, USA 
  • Tao Mei, Microsoft Research Asia, China 
  • Jiebo Luo, University of Rochester, USA
Comments