Group on Interactive Coding of Images
Group on Interactive Coding of Images (GICI) Department of Information and Communications Engineering (DEIC) Universitat Autònoma de Barcelona (UAB)
About us
Presentation

The Group on Interactive Coding of Images (GICI) focuses on the study, design and implementation of image coding and transmission techniques and standards.

GICI collaborates with other national and international research groups and is open to new collaborations with other researchers. Students can also do their Master and Ph.D. theses with us.

GICI belongs to the Department of Information and Communications Engineering (DEIC) at the Universitat Autònoma de Barcelona (UAB).

Our research

The GICI group continuously struggles to push the state of the art of image coding and transmission. As a result, research papers and software tools are published. The main lines of research that are currently being developed are the following:

Remote sensing

Images captured by sensors located in satellites or airborne ships have been studied in our group. Remote sensing images include hyperspectral and cartographic data, which have special features that can be exploited to obtain better compression factors.

Some of these images have to be compressed onboard, which severely restricts the available memory and CPU usage. In addition, the large dimensions of cartographical data motivates the development of interactive transmission protocols.

Biomedical images

Biological and medical images used for clinical and research purposes are of special interest to the GICI group.

Images produced by Computer Tomography (CT), micro-CT, X-Ray, DNA microarray experiments and Whole-Slide Images have been researched. They exhibit properties very different from natural images and are particularly hard to code. Several proposals have been already published to improve the compression performance for this type of images, and more are being currently investigated.

Bitplane image coding

The bitplane image coding strategy is employed by some of the most successful image coders, including JPEG2000, the latest image compression standard.

Several original strategies concerning the scanning order and embedded quantization have been published. The efficient paralellization of bitplane coders is also being investigated with the goal of using GPU architectures.

Contact

If you are further interested in any aspect of our group, please do not hesitate to contact us (gici-info@deic.uab.cat).