The research of the Images and Visual Representation Group (IVRG) in the Laboratory of Audiovisual Communication (LCAV) is primarily concerned with color image processing for photography and video to optimize the quality of natural image encoding and display. The developed image algorithms and models are based on low and high level processing features of the human visual system.

In general, we are interested in the question of "what is an image?" What are the fundamental components that "make" an image, and how does the change of one of those components influence the perception of it? How come we always know if we are looking at a "good" image or a "bad" image, even without being able to identify what about it is "good" or "bad?" Can we identify models and algorithms that define and predict these subjective criteria? Especially in the area of color imaging, there are fundamental gaps in our knowledge of what color - or a change in color - contributes to the perception of an image. Most color imaging models and algorithms are based on single pixel transformations and measures, even though it is a well-known fact that the appearance of color is directly influenced by its size, surround, and viewing conditions. Adding color to image processing tasks introduces multidimensionality beyond individual processing in three channels.

To develop meaningful color image processing and image quality models and algorithms based on visual perception needs the collaboration of several fields, including photography, computer vision, image processing, psychology, and physiology. IVRG is therefore collaborating with several experts from different fields and different institutions. Applications of the developed methods and algorithms are in the cross-comparisons of different image capture, processing and display techniques, in color management workflows, and in digital image production and archiving.

Research topics

Assessing human skin color from uncalibrated images
HDR CFA Image Rendering
Frequency-tuned salient Region Detection
Illuminant retrieval for fixed location cameras
Color assessment for Home Décor
Shadow detection and removal
Near-Infrared Imaging
Image re-targeting
Database Saliency and Image Summaries
Portrait enhancement using NIR at the Swiss Camera Museum
Superpixel segmentation

Research Support

IVRG's research is supported by the Swiss National Science Foundation.
IVRG is a member of the National Competence Center in Research on Mobile Information and Communication Systems (NCCR-MICS), a center supported by the Swiss National Science Foundation under grant number 5005-67322.
IVRG participates in PHAROS (Platform for search of Audiovisual Resources across Online Spaces), a EU integrated project (FP6)
IVRG is a member of K-Space, the European Network of Excellence in Knowledge Space of semantic inference for automatic annotation and retrieval of multimedia content.
IVRG is also supported by Hewlett-Packard Labs and Logitech.

Past research topics

Automatic Red-Eye Removal
Chromatic Adaptation
Color Transparency
Dynamic Texture Synthesis
High dynamic range image rendering
Image Classification Using Content And Context Information
Joint superresolution and demosaicing
Low-level representation of visual features
Online course on scientific and forensic photography
Perceptual Video Quality
Reading Text Appearing in Natural Images
Salient Region Detection and Segmentation
Superresolution imaging
The Flux: creating a large image database
Virtual Camera