Call for Papers: ICCV-98 Workshop
Conceptual Description of Images
Workshop on Conceptual Descriptions of Images
2 January 1998
In conjunction with
International Conference on Computer Vision (ICCV-98)
This is the second workshop on Conceptual Description of Images,
following the one in April 1996 at Cambridge, UK held in conjunction
|Elisabeth Andre (DFKI, Germany)
|Mike Brady (Oxford, UK)
|Aaron Bobick (MIT, USA)
|Christopher Brown (Rochester, USA)
|Hilary Buxton (Sussex, UK)
|Tony Cohn (Leeds, UK)
|David Hogg (Leeds, UK)
|John Kender (Columbia, USA)
|Paul Mc Kevitt (Sheffield, UK/Aalborg,Denmark)
|Amitabha Mukerjee (Kanpur, India)
|Hans-Hellmut Nagel (Karlsruhe, Germany)
|Patrick Olivier (Aberystwyth, UK)
People often use non-detailed descriptions of visual scenes that may be
called "conceptual". Such information is very useful, since it compactly
describes the properties of the visual objects and their inter-relations, the
task context, etc. Visual applications that may benefit from such
descriptions include, for example, visual surveillance, document
processing, multimedia and GIS, automobile driving assistants, etc.
The workshop aims to stimulate collaborative research between the
fields of research which impinge on this problem, such as
Computer Vision, Natural Language Processing, Spatio-Temporal
Reasoning, Cognitive Science, Computational Imagery, Linguistics, etc.
Papers should address the goal of deriving and using meaningful
descriptions of the scene, the task and the context.
Research topics that address the issues of representing visual knowledge
and processing images to obtain conceptual descriptions include:
- Context-based vision.
- Integration of vision and natural language.
- Spatial and temporal reasoning in images.
- Representation and control of visual behaviours.
- Inferring agent behaviours/goals from image sequences.
Attendance and Workshop Format
It is intended that a diverse group of researchers attend the workshop.
Registration in the main ICCV conference is not mandatory.
The format will have more discussion sessions, each session of 3
papers followed by a mini-panel on the session theme, providing the
audience a larger role, and speakers can also have a wider discussion
following their work. Participants who do not wish to present papers
are requested to submit a writeup addressing the following points:
- How does your work differ from, or extend existing
approaches? What are the limitations and assumptions of
- How should domain knowledge be represented?
- What are the open questions?
Proceedings with all accepted papers will be distributed at the
workshop. Also, a special issue of the Journal Image and Vision
Computing is planned on this topic, for which some of the authors
will be invited to submit extended versions of their papers.
Papers must be a maximum of 30 pages including title, abstract, and
author affiliations. Four copies should be mailed to either address
below. Electronic submission is also encouraged (either postscript or
Cognitive and Computing Sciences
University of Sussex
Falmer, Brighton, BN1 9QH, UK
Tel: +44 1273 678569
Center for Robotics
Indian Institute of Technology
Kanpur 208016 INDIA
|Submit Full Papers by: || September 15, 1997 |
|Acceptance Notification by: || October 30, 1997 |
| Camera-ready papers due: || November 30, 1997
The venue of the ICCV conference as well as this workshop is the city
of Bombay on the Arabian Sea. Bombay, the business capital of India,
is directly connected to major cities in the world and India. Nearby
tourist attractions include the world famous cave paintings and
sculptures of Ajanta and Ellora.
Sea cruises to several islands and the resort of Goa are possible.
Amitabha Mukerjee, Fri Wed May 14
12:28:00 GMT+0500 1997
[ICCV-98 HOME PAGE]