Special Session on Integrative Computer Vision and Multimedia Systems

Important information

• Submission deadline: December 15, 2017 February 2, 2018 (extended)
• Acceptance notification: March 16, 2018
• Camera ready deadline: April 19, 2018
• Online submission: https://submissions.mmsys2018.org/computervision
• Submission format: 6-12 pages, using ACM style format (double-blind). Please see the submission guidelines for more information about the process.
• Reproducibility: obtain an ACM reproducibility badge by making datasets and code available (authors will be contacted to make their artifacts available after paper acceptance)

Special Session chairs

• Ketan Mayer-Patel, University of North Carolina, USA
• Wu-Chi Feng, Portland State University, USA

Scope & goals

There is an increasing overlap between the computer vision research community and the multimedia systems research community driven by real-world vision-driven applications such as autonomous vehicles, teleimmersion, and augmented reality in which application-level semantics of streaming video sources plays a significant role. Going forward mulitmedia systems will increasingly need to integrate content-based decisions such as computer vision algorithms into their core.  This research will need to explore bringing traditionally heavy-weight computer vision algorithms into the smallest devices (e.g., IoT / sensors) and will need to effectively trade off time, video quality, bitrate, and / or frame rate into vision algorithm design.

At the same time, multimedia systems research and development will need to break free of the assumption of human consumption that lies at the heart of the current state-of-the-art media representations and streaming protocols. Vision-driven feedback about saliency and semantics will need to be incorporated into multimedia system adaptation and control mechanisms. 

This special session on computer vision multimedia systems is intended to help bridge the gap between the computer vision and multimedia system research communities by providing a forum for high-quality research papers that explore challenges faced in this overlap. The special session is an opportunity to foster increased dialogue between these research communities and encourage new and exciting avenues of research that span across them.

Topics of interest for this special session include, but are not limited to:

• Real-time adaptive computer vision algorithms
• Compression-aware computer vision
• Vision-based compression
• Media streaming in support of computer vision
• Compressed-domain vision processiong
• Real-world computer vision applications
• Stereoscopic, 3D, and video+depth streaming
• Multimedia support for structure from motion and 3D modelling
• Vision-driven surveillance and monitoring systems
• Video segmentation and understanding grounded by computer vision
• Computer vision in human computer interaction
• Multimedia visual content representation and understanding
• Computer-vision with multiview video streaming
• Real-time event detection and understanding
• Saliency and region of interest detection and tracking
• Intelligent vision systems
• Frameworks for vision-driven multimedia system development and prototyping

Papers should be between six and twelve pages long (in PDF format) including references prepared in the ACM style and written in English. Hence MMSys papers enable authors to present entire multimedia systems or research work that builds on considerable amounts of earlier work in a self-contained manner. MMSys papers are published in the ACM Digital Library; they are available for just as long as journal papers and authors should not feel compelled by space limitations to publish extended works on top of an MMSys paper. Authors who submit very specific, detailed research work are encouraged to use less than 12 pages. The papers are double-blind reviewed.