Image-Cube

VisiQ

Computational Salience Based Approach for Print Quality Assessment

The VisiQ is developing an unsupervised method to categorize an unknown set of images into seperate clusters.

Goals

The goal of the project is to develop a system that is able to categorize a set of unknown images. Images should be categorized based on objects that images consist of. The system must be unsupervised, hence it cannot have any prior information about the images, e.g. class labels, training images, etc.

News

26.4.2010
Randomized Caltech 101 image set published and added to the webpage. Please see the downloads section.
19.3.2009
Page updated. New links added to a image database and rival systems.
25.3.2009
Unsupervised VOC www-demo published. Try! Limited access.
2.8.2008
Website published.

Contact information

Lappeenranta University of Technology
Teemu Kinnunen WWW Doctoral student, Master of science
Joni Kämäräinen WWW Professor
Lasse Lensu WWW Associate Professor
Heikki Kälviäinen WWW Professor
Helsinki University of Technology
Mari Laine-Hernandez not available Doctoral student, Master of science
Pirkko Oittinen not available Professor

This project is part of the research in the Machine Vision and Pattern Recognition Research Group.

Downloads

Demo

Demo is based on the system that was developed for Kinnunen's Master's thesis Unsupervised visual object categorization, (2008).

WWW-DEMO (Limited access)

Image search engines

Databases and competitions

Researchers

Image categorization systems

Interest points (detectors and descriptors)

Saliency

Publications

Do not hesitate to contact authors (e-mail, etc.) in order to retrieve copies or reprints of the following publications.