Course content

This course covers advanced topics and methods within selected areas of multimodal visual data analysis and processing. It contains three main parts.

First a set of core topics are covered in depth through lectures and compulsory guided practical exercises. This part will be evaluated by a mid term exam. Core topics include but are not limited to:

• Advanced methods for image and video quality assessment
• Multi- and hyperspectral image acquisition, processing, and analysis
• Processing and analysis of 3D data
• Advanced multimodal and multidimensional visual data processing (data fusion, pan-sharpening, super-resolution)
• Measuring, modeling and reproducing material appearance

In a second part the students will be assigned recent research papers which they will review, present, and discuss with each other and the involved instructors. Also invited guest lectures on selected research topics within the field will take place in this part, focusing on visual informatics application areas such as medical imaging, remote sensing, cultural heritage, and machine vision. A compulsory reflection note on learning outcomes will be delivered and evaluated based on this part. The specific topics covered in this part will vary from year to year.

The last part is an individual project work, on a topic selected by the student and approved by the instructor. Evaluated through a written report, oral presentation, and examination.

The course emphasises reproducible workflows and hands-on analysis using real-world datasets and research-grade tools.

Learning outcome

Knowledge

Students will be able to:

• Demonstrate specialized insight and advanced knowledge within the broadly defined field of visual informatics, including methodologies of capturing, processing, assessing, and reproducing multimodal images
• Demonstrate specialized insight and good understanding of the research frontier in a selected area of visual informatics
• Discuss domain-specific requirements and solutions in application domains for visual informatics such as n medical imaging, remote sensing, cultural heritage and industrial vision.

Skills

Students will be able to:

• Process and analyse multimodal visual datasets using suitable software tools.
• Implement and evaluate quality assessment methods for imaging systems.
• Analyze existing theories, methods, and interpretations, and to challenge established knowledge and practice in the visual informatics field
• Design and execute a research-oriented project in the visual informatics domain

General competence

Upon completing the course, students will:

• Analyse and justify design choices in multimodal imaging and visual informatics pipelines.
• Communicate results and limitations clearly to specialists and non-specialists.
• Reflect on ethical, societal and domain-specific aspects of visual informatics applications.

Learning methods and activities

Learning will take place through lectures, seminars and practical assignments, including exercises and projects. Some sessions may involve group discussions or guest lectures on related research topics.​

Compulsory requirements: Completed exercises

Teaching Materials:

-Hand-outs and research papers

Compulsory assignments

• Reflection note

Specific conditions

Recommended previous knowledge

Basics of Digital Information Processing, Visual Informatics and Machine Learning

Course materials

Hand-outs and research papers