Special Issue on Semantic Multimedia Data
Future Generation Computer Systems
ISSN: 0167-739X, Elsevier
Multimedia data can be easily uploaded, communicated and shared in community portals. These later allow users to manually tag, comment and annotate the digital content, but they lack a general support for fine-grained semantic descriptions and look-up, especially when talking about things “inside” multimedia content, such as an object in a video or a person depicted in a still image.
Linking Multimedia Data is an active and inter-disciplinary research field in multimedia. Turning a distributed repository of images, video, text, and other signal-based objects (such as various radar signatures) into an easily browsable information and knowledge would certainly transform the underlying data into a more satisfying representation for humans to intelligently navigate. The power of this approach results from the mixture of different modalities of data. While linked textual data is being studied by the web semantics and database communities, adding non-textual data is much more satisfying for human interaction, but comes with the price of more complexity.
While in hypermedia, one mainly focuses on languages and synchronization made between information parts in both temporal and spatial dimensions, we mainly focus in this special issue on the (semi) automatic and collaborative methods for fostering semantics in multimedia data by enriching information (so to come up with a knowledge), for visualization and exploration, and for missing (meta)data estimation. The added complexity is due to objects having multiple facets, depending on their use and who or what system is doing the labeling.
This research area is a blending of topics of interest to many disparate research communities. The novelty will reside initially in how to formulate and implement the boundaries between tasks of interest to these different communities. As the field matures and the diverse nature of the data is accepted as a given by researchers in this area, research advances of a more integrative nature will be the norm, treading into areas that we can only dream of today.
Topics of interest include, but are not limited to:
• Multimedia data linking methods
• Collective knowledge, event and entity detection
• Multimedia measures (similarity, quality, data fusion, etc.)
• Semi-supervised, learning-based multimedia data linking methods
• Multimedia on the web sampling techniques
• Identity representation and semantics
• Online personal information management
• Reasoning on connected multimedia data
• Retrieval and annotation of big multimedia data
• Big multimedia data analytics
• Semantics and metadata in multimedia systems
• Link propagation
• Provenance and trust models on links
• Semantic-based multimedia retrieval and annotation
• Emergent semantics in content-based retrieval systems
• Social media and Crowdsourcing
• Evaluation techniques and tools
• Marketplaces, aggregators and indexes for Semantic Multimedia Data
o digital humanities
o social networks
o Internet of things
Manuscript Due: January 15, 2018
First Notification: June 15, 2018
Revised Manuscript: August 1, 2018
Notification of Acceptance: October 15, 2018
Camera Ready Paper Due: November 30, 2018
Publication Date: Last Quarter of 2018
Relevant best papers presented at SITIS 2017 (the 13th IEEE International Conference on Signal Image Technology and Internet Based Systems) can be extended and submitted to the special issue. The extended versions must contain at least 30% of new materials and include results related to the major topics of interest.
Submissions should be prepared and submitted according to the author instructions available at the journal homepage, https://www.journals.elsevier.com/future-generation-computer-systems/
- Richard Chbeir, University of Pau and Adour Countries, France ()
- Albert Dipanda, University of Bourgogne, France ()
- Luigi Gallo, National Research Council of Italy, Italy ()
- Francesco Piccialli, University of Naples Federico II, Italy (