Welcome to the Institute of Computational Perception!

We are a (still) young and growing institute with innovative projects and motivated young researchers. The focus of our research and teaching is on Artificial Intelligence and Machine Learning.

We develop and study computational models and algorithms that permit computers to perceive and 'understand' aspects of the external world, where we interpret 'perception' in the widest sense of the word, as the extraction of useful high-level information and knowledge from complex, possibly low-level data (audio, video, images, sensor data, texts, or even the Internet).

This requires fundamental and applied research on AI, machine learning, pattern recognition, text / data / social media mining, signal processing, statistical data modeling and classification.

Our current research has a particular focus on intelligent audio and music processing. Further research topics include biometrics, cryptography, personalization, and recommender systems.

Our goal is to offer state-of-the-art research and teaching in this area, and to provide a teaching environment that permits students to get involved in real research projects as early as possible.


Logo Institue of Computational Perception


Johannes Kepler Universität Linz
Altenberger Straße 69
4040 Linz


Science Park 3, 4th floor
Aubrunnerweg 3B/1


Mon-Thu: 07.30 - 15.00


+43 732 2468 4700

News & Events
News 25.01.2021

Keynote presentation at ICAART 2021

Gerhard Widmer will give a keynote presentation at the 13th International Conference on Agents and Artificial Intelligence (ICAART 2021).

News 15.01.2021

How to break out of your Spotify feedback loop and find new music

Article based on interviews of CP member Markus Schedl and former CP member Peter Knees on diversification of Spotify music recommendations published in Wired UK.

News 08.01.2021

Gerhard Widmer has been elected as an ELLIS Fellow by the European Laboratory for Learning and Intelligent Systems (ELLIS)

Ellis Fellows advance science and act as ambassadors of ELLIS. They provide strategic advice and leadership not just scientifically, but also in terms of how to build and grow ELLIS.

News 11.11.2020

CP submission wins the "Judges Award" in DCASE 2020 Challenge

"Low-Complexity Cross-Device Acoustic Scene Classification with RF-Regularized CNNs" by Khaled Koutini, Florian Henkel, Hamid Eghbal-zadeh and Gerhard Widmer win the "Judges Award" in DCASE 2020 Challenge.

View here