Site visit in Oulu on 31 March 2003

The coordination unit made a site visit at the Machine Vision Group at the University of Oulu on 31 March 2003. Professors Matti Pietikäinen and Olli Silvén together with their research group presented their research in machine vision and expecially their PROACT project Machine Vision for Sensing and Understanding Human Actions.

The objective of the project is to investigate the capabilities of machine vision in proactive computing and to develop solutions needed for building emerging applications. The Machine Vision Group already has achieved several promising results in the field of machine vision. For example, the group has developed methods for skin detection under varying illuminations conditions. A facial image database has been collected to support this research.

The group has also developed methods for detecting and recognizing faces. The problem can be stated as follows: Given still or video images of a scene, identify or verify one or more persons in the scene using a stored database of faces. The solution to the problem consists of solving several subtasks in a sequential manner: face detection, face normalization, and face recognition. To detect faces the group uses color-based methods based on skin locus and successive detectors.

The group also studies face recognition in varying environments. Several demos, both of face detection, face tracking and face recognition, are presented at the group's web site, http://www.ee.oulu.fi/mvg/mvg.php. Other topics related to face analysis that will be studied are facial feature extraction, tracking of head movements and use of head movements for human computer interaction.

The group has research activity in  tracking humans in varying conditions.  Motion is one of the primary cues for detecting targets from the viewed scene. An analysis technique has wider applicability if it can also deal with videos taken with a moving camera. Several demos of human tracking are also available at the group's web site. With the help of the trajectories of the moving objects and the velocity, the methods are able to count the total number of people passing and to distinguish, for example, between pedestrians and cyclists.

The goal of the PROACT project is to develop an embedded vision module using novel vision software architecture. It will be provided with machine vision skills for sensing and identifying humans and their actions in varying conditions and with capabilities to wirelessly communicate with other units. With such vision sensors, proactive systems can sense the movements of objects, detect and track people or other objects, personalize services by recognizing people, recognize gestures, text and other objects or events, create a 3-dimensional model of the environment, etc. The group also intends to demonstrate different types of proactive functions using multiple cooperating vision modules.

More information
For more information, please contact Professors Pietikäinen and Silvén at the University of Oulu or see the home page of the group at http://www.ee.oulu.fi/mvg/mvg.php. See also the project's web page at this site.

Researchers involved in the project are

• University of Oulu, Department of Electrical and Information Engineering, Information Processing Laboratory, Machine Vision Group:
• Professors Matti Pietikäinen and Olli Silvén, Dr. Janne Heikkilä
• Researchers Abdenour Hadid, Eric Galloix
• Assistant researchers Jari Hannuksela, Pentti Saarinen

Greger Lindén
Programme Coordinator
Greger.Linden@cs.helsinki.fi