Key Challenges for Multimedia Research in the Next Ten Years

The multimedia and multimodal (MM) research community is dedicated to research on the acquisition, communication, analysis, modeling, interface design, and impact of applying multimodal-multimedia data to challenging problems such as semantic information extraction, co-processing and interpretation of multiple heterogeneous modalities, seamless humanmachine interaction, and scalable distributed collaboration. This community of researchers also aims to advance important applications such as education, healthcare, advanced communications and social networking. One of the key objectives for MM research is to model and, in some cases, exceed humans’ ability to process multi-sensory information during activities like communication and learning. The multidisciplinary MM field is distinct from other research areas in its emphasis on deep, meaningful integration of multimodal data to enhance the reliability and depth of information derived, as well as the quality of multimedia user experience. A major advantage of fusion-based multimodal interaction and systems is greater robustness and coverage of information processing and improved experience of interaction, which typically exceed that possible based on the sum of individual modalities. And, the synergy between multimedia-multimodal data sources is often the prime source of generating new knowledge.The MM community has demonstrated extremely productive activities in the last two to three decades. Starting with the inaugural ACM Conference in Multimedia (ACMMM) in 1993, the ACM International Conference on Multimodal Interfaces (ICMI) in 2003, and high-quality …