MHC 05

ACM Workshop on Multimedia for Human Communication - From Capture to Convey

Hilton, Singapore, November 11 2005
in conjunction with ACM Multimedia 2005

wave japanese style

Important Dates

November 11th MHC 05 Workshop at the Hilton Hotel Singapore, Indonesian Room at Level 5.


8:30 - 8:45 Welcome
8:45 - 9:15 Session Models
Lynda Hardman: Canonical Processes of Media Production
Alejandro Jaimes: A Component-based Multimedia Model
9:15 - 10:00 Session Applications
Brigitte Kerhervé: Design and Production of New Media Artworks
Frank Nack: Capture and Transfer of Metadata During Video Production
Nicholas Diakopoulos: Supporting Personal Media Authoring
10:00 - 10:30 Morning Tea (Foyer)
10:30 - 12:30 Discussion and modeling
12:30 - 13:30 Lunch (Restaurant, 24th Floor)
13:30 - 15:30 Modeling
15:30 - 16:00 Afternoon tea(Foyer)
16:80 - 10:30 Final modeling and wrap-up discussion


This workshop is organized as a brainstorming session. Participants are requested to get involved before, during and after the meeting.

All participants should familiarize themselves with the relevant ideas of the papers as provided in the schedule above and extract a list of key questions. Please send these questions to so that the clustered questions can be taken as corner stones for the discussion during the workshop.

Moreover, a wicki is established to exchange ideas before the meeting. Participants are strongly encouraged to participate in the collaborative edit of this document once available, as it will serve to normalize the terminology.

During the workshop participants with accepted statements will have time to present them briefly (15 minutes each). The statements will also be provided as posters to be consulted during the discussions. The essential part of the time will be devoted to discussion.

The organizers will prepare a workshop report containing the results of the workshop as well as the next steps in development. Participants are encouraged to contribute to this report. The position papers and this report will be made available in the ACM SIGMM website. A publication in form of either a journal article or a special issue in one of the leading journals (ACM TOMCCAP or IEEE MM) is anticipated. We will also establish a Wikipedia entry about the "Multimedia communication model".


Frank Nack, CWI, NL

Program Committee

Chitra Dorai, IBM Watson, USA
Lynda Hardman, CWI, NL
Brigitte Kerherve, Université de Quebec a Montréal, Canada
Stephen Kimani, Università di Roma, Italy
Frank Nack, CWI, NL
Kurt Piersol, Ricoh Innovations, USA
Nicu Sebe , UvA, NL
Freddy Snijder, Philips Research, NL

Workshop Scope

In recent years we have seen a shift in multimedia computing from system aspects and related low-level feature research towards the understanding of high-level media semantics as illustrated by research on bridging the semantic gap or experiential systems. The aim of these research directions is to facilitate new forms of creativity, knowledge exploration and social relationships, mediated through communication networks. The essence of such systems is to allow humans to communicate better with each other mediated through technology. The problem, however, is that today's multimedia processing systems do not take a holistic view on multimedia communication. This results in multimedia environments that establish artificial borders between the role of the user as consumer and creator of media, hindering the use of multimedia for human communication by a wide audience. Our vision is that the multimedia computing community should create technical solutions that can be combined together as different stages of a dynamic whole.

The challenge is to develop systems that understand the role they play in the communication process chain and allow alternatives for mixing and matching defined sets of functionality which are able to blend into the overall context the user operates in. The task of the technology is to support the essential building blocks of media creation and manipulation: such as capture, annotate, archive, select, organise, present and convey. Multimedia environments of the future have to be fluid, where humans or machines can carry out tasks on an equal basis. That requires that we, as a community, model the inputs and output of the separate processes sufficiently well so that system builders are able to define the inputs and outputs in terms of this model.

The goal of this workshop is to work towards an initial formulation of a model, namely to identify commonly carried-out tasks in media creation, manipulation and organization and consumption and to create preliminary descriptions of input and output specifications. The aim of this workshop is to establish the task set and possible paths among tasks, and gain a better understanding of the inputs and outputs for the identified tasks as well as their inner process structure.