Image Compression

Image Compression

Description

Motivation: High resolution image data is becoming increasingly popular in research and commercial applications (e.g. entertainment, medical imaging). In addition, there is also a high demand for content distribution via the internet. Due to the resulting increase in storage and bandwith requirements, image compression is a highly relevant and very active area of research.

Teaching Goals: The course is designed as a supplement for image processing lectures, to be attended before, after or parallel to them. After the lecture, participants should understand the theoretical foundations of image compression and be familiar with a wide range of classical and contemporary compression methods.

Contents: The lecture spans the whole evolution of image compression from the dawn of information theory to recent machine-learning approaches. It is seperated into two parts:

The first half of the lecture deals with lossless image compression. We discuss the information theoretic background of so-called entropy coders (e.g. Huffman-coding, arithmetic coding, ...), talk about dictionary methods (e.g. LZW), and cover state-of-the-art approaches like PPM and PAQ. These tools are not limited to compressing image data, but also form core parts of general data compression software such as BZIP2. Knowledge about entropy coding and prediction is key for understanding the classic and contemporary lossless codecs like PNG, gif or JBIG.

The second part of the lecture is dedicated to lossy image compression techniques. We deal with classic transformation based compression (JPEG, JPEG2000), but also with emerging approaches like inpainting-based, fractal, or neural network compression. Furthermore, we consider related topics like human perception, and error measures.

 

Entrance Requirements

Basic mathematics courses (such as Mathematik für Informatiker I-III) are recommended. Understanding English is necessary. Image processing lectures such as "Image Processing and Computer Vision" are helpful for some specific topics, but not necessary. For the programming assignments, some elementary programming skills are required.

 

Assessments / Exams

There will be two written exams with a limited open book format. Concrete dates will be announced before the start of the semester.

Detailed rules for our exams are published in the self test assignment which also provides an impression of the structure and assignment types to be expected from the real exams.

You can participate in both exams, and the better grades counts. Please remember that you have to register online for the exam in the LSF system of Saarland University.

If you cannot attend the exam, contact Pascal Peter as early as possible. In case you have proof that you cannot take part for medical reasons or you have another exam on the same day, we can offer you an oral exam as a replacement. Note that we need written proof (e.g. a certificate from a physician/Krankenschein) for the exact date of the exam.

Lecture Format

More details on the lecture format will be available before the beginning of the semester.
 

Lecture Materials / Assignments

All slides, assignments, example solutions, and an exam formulary are offered for download in the CMS.

 

References

There is no specific book that covers the complete content of this class. However, each of the following books covers several of the topics discussed in the lecture:

T. Strutz: Bilddatenkompression. Vieweg+Teubner (in German)

D. Hankerson, G. A. Harris, and P. D. Johnson, Jr.: Introduction to Information Theory and Data Compression. Chapman & Hall/CRC

K. Sayood: Introduction to Data Compression. Morgan Kaufmann

Further references will be given during the lecture.