Course content

The course addresses advanced and emerging topics in visual information processing, including but not limited to:

• Image and Video Processing and Enhancement: Advanced algorithms for denoising, super-resolution, restoration, and perceptual quality improvement.
• Object Detection and Recognition: Techniques for identifying and classifying objects in images and video streams using traditional and deep learning-based approaches.
• Scene Understanding and Navigation: Methods for spatial reasoning, path planning, and autonomous navigation in complex environments.
• Quality Assessment and Evaluation: Objective and subjective metrics for evaluating visual fidelity and performance of processing algorithms.
• Spectral and Multispectral Imaging: Acquisition and processing of spectral data for applications in material analysis, medical imaging, and remote sensing.
• 3D and Volumetric Data Representation: Techniques for modeling, visualization, and interaction with three-dimensional and volumetric datasets.
• Machine Learning and AI for Visual Data: Deep learning architectures for segmentation, tracking, and pattern recognition in visual datasets.
• Human Visual Perception Models: Incorporating perceptual principles into algorithm design for improved realism and usability.

Learning outcome

Knowledge (Remembering, Understanding, Analyzing)

• Remember key concepts and terminology in visual information processing, including image representation, perception models, and computational techniques.
• Understand theoretical foundations and principles behind advanced visual information processing methods.
• Analyze the appropriateness and applicability of different algorithms and models for solving complex visual processing tasks.

Skills (Applying, Analyzing, Creating)

• Apply advanced visual information processing techniques to real-world problems such as image enhancement, segmentation, and object recognition.
• Analyze performance and limitations of various algorithms in interdisciplinary applications.
• Create and implement optimized solutions and algorithms for visual information processing tasks using modern tools and frameworks.

General Competence (Evaluating, Creating)

• Evaluate recent research contributions and emerging trends in visual information processing critically and constructively.
• Create well-structured scientific reports and presentations that communicate complex ideas effectively to both technical and non-technical audiences.
• Collaborate and provide informed feedback on others’ work in visual information processing, demonstrating academic integrity and professional standards.

Learning methods and activities

The course will adopt a blended learning approach, combining traditional lectures, expert seminars, and student-led presentations.

• Lectures: Delivered by the course responsible and invited experts, covering fundamental and advanced topics in visual information processing.
• Seminars: Interactive sessions where students present and discuss assigned topics, fostering critical thinking and collaborative learning.
• Project Work: Students will work individually or in groups on research-oriented assignments. Each group or individual will be assigned a supervisor from the course teaching team to provide guidance throughout the project.

All lectures, seminars, and student presentations are considered integral parts of the course and contribute to achieving the learning outcomes.

Specific conditions

Recommended previous knowledge

A solid foundation in visual information processing and analysis is expected. Relevant prior coursework may include:

• Image and Video Processing: Techniques for enhancement, filtering, compression, and feature extraction.
• 3D Information Representation and Analysis: Methods for modeling, reconstruction, and interpretation of three-dimensional data.
• Volumetric Data Visualization and Interaction: Approaches for rendering, exploring, and interacting with complex volumetric datasets.
• Additional topics such as computer vision, pattern recognition, and machine learning for visual data are highly beneficial.

Required previous knowledge

Participants are expected to have:

• Basic knowledge of image processing: Understanding of fundamental concepts such as filtering, enhancement, and feature extraction.
• Programming skills: Ability to write and debug code in a relevant programming language (e.g., Python, C++, or MATLAB).
• Foundations in algorithms and data structures: Knowledge of core principles for efficient computation and problem-solving.
• Introductory image processing experience: Familiarity with basic techniques for handling and analyzing visual data.

Additional experience in computer vision, machine learning, or visualization techniques is advantageous but not mandatory.

Course materials

Recent research papers, online tutorial/GitHub links, and lecture slides. Research papers and other relevant teaching material used in the seminars will be made available electronically.