Intelligent Systems Lab Amsterdam

Intelligent Autonomous Systems perceive their environment through sensors for goal-directed actions even in dynamic situations.


Perception

Learning

Geometric principles

Multiagent systems

Information and Language Processing Systems analyze text using the structure of documents to reveal semantic and emotional meanings.


Information retrieval

Language technology

Semistructured data
 

Intelligent Sensory Information Systems create access to the content of digital images and video.

color
Features
forensics
Multimedia access
cognitive vision
Cognitive vision
forensics
Image semantics

The Intelligent Systems Lab Amsterdam ISLA at the University of Amsterdam performs fundamental, applied and spin-off research. We define intelligence as observing and learning; observing the world by video, still pictures, signals and text and abstracting knowledge or decisions to act from these observations.

At ISLA we prefer to study hard scientific problems from real data with real applications. We typically analyze visual or textual data derived from video repositories, the Internet, or any other kind of sensory data: search engine logs, feeds, hand-held video recordings, mobile robot observations and so on all in order to understand its content.

Successful applications have been achieved in video search engines, delivering one of the top performers in the international competition for video search engines amidst competition from the US and China. We have also performed best in the international robot soccer simulation competition testing decision cooperation between autonomous systems. We have performed best in the competition for query-answer based search engines for large text repositories. ISLA's mood analyzer estimates the mood of the world on-line by analyzing just updated blogs anywhere in the world, and its emotion recognizer estimates human expression of emotion. See Research applications for more fundamental or applied science examples.

ISLA is composed of three groups. Intelligent Autonomous Systems, headed by prof. Frans Groen, perceive their environment through sensors for goal-directed actions even in dynamic situations. Information and Language Processing Systems, prof. Maarten de Rijke, analyze text using the structure of documents to discover and track actionable meaning. Intelligent Sensory Information Systems, prof. Arnold Smeulders, create access to the content of digital images and video.

ISLA is one of the largest labs of its kind in the world. It is part of the University of Amsterdam, and funded by NWO, EU-grants, companies, as well as large public-private grants enhancing the cooperation between universities and companies. ISLA's structural partners are TNO in the Centre for Intelligent Observation Systems, the Dutch Forensic Institute, Beeld & Geluid for video search engines, IlseMedia for text and image searching and browsing, and the HvA Almere on domotica. ISLA plays a leading role in the Amsterdam Centre for Creative Content and Technology.

Maintained by Michiel. Last edited on Mon, 25 Jan 2010 13:39:11 +0100