HMMP 2015 - Workshop on Human Memory-Inspired Multimedia Organization and Preservation (HMMP)

With the growing volume, diversity and temporal dynamics of multimedia content on the Web, in social sharing platforms and in personal and institutional archives, there is a clear need for new approaches for efficiently organizing multimedia content and making it easily discoverable and accessible; and, preserving it discoverable, accessible and understandable in the near as well as distant future. This necessitates, among many other things, keeping the multimedia information in sync with the time-evolving context information that is needed for its interpretation at different times. The above are by no means easy tasks for computing machines.
Humans are very effective in remembering and organizing information by abstraction, pattern exploitation, or contextualization; and, significant progress has been made in recent years towards endowing computing machines with the algorithmic tools that are needed for performing similar functions on large volumes of multimedia in an automated or semi-automated fashion (e.g., progress in machine learning for multimedia). Besides these functions, though, humans are also capable of forgetting irrelevant details; an important function of the human brain that helps us to focus on relevant things instead of drowning in details by trying to remember everything.
The research question that the HMMP’15 workshop aims to address is: can we be inspired from the human remembering and forgetting processes for developing more advanced algorithms for multimedia organization (incl. search, retrieval, annotation, summarization) and preservation (incl. contextualization)? In particular, we are interested to know how human remembering and forgetting can play a role in multimedia organization and retrieval, in applications such as search in personal and organizational information spaces, as well as in the social multimedia Web; and how the effects of time in human remembering and forgetting can help us to maintain multimedia content discoverable, accessible and understandable (contextualized) in the near and distant future, thus supporting its long-term preservation.
This workshop aims to bring together researchers and practitioners from the different fields that contribute to the development of technologies and applications for multimedia organization (search, retrieval, annotation, summarization) and preservation (contextualization); and, get them in touch with researchers from a different scientific field that investigates how humans organize, remember and forget information.
Human memory-inspired approaches to multimedia search and retrieval
Topics of interest to the workshop include but are not limited to:
Human memory-inspired approaches to multimedia search and retrieval
Time-aware approaches to multimedia search and retrieval
Human memory-inspired approaches to multimedia annotation
Human memory in language processing and machine learning
Analyzing societal memory or collective memory in the social multimedia Web
Memory-inspired multimedia summarization
Time-aware multimedia contextualization for long-term remembering and reminiscence
Long-term multimedia management and access, e.g. lifelogging
Multimedia system design to support human memory from the Human-Computer Interaction (HCI) perspective
Other novel multimedia organization and preservation applications