Multimedia Analytics

Course manual 2024/2025

Course content

Multimedia analytics aims to develop AI techniques for getting the richest information possible from the data (image/video/text/graphs), interactions surpassing man and machine intelligence, and visualizations blending it all in effective interfaces. In this course we study both the algorithms underlying multimedia analytics and take a holistic view to bring all the components together in innovative multimedia analytics solutions. An important part of the course is a large multimedia analytics group project resulting in a demonstrator and a scientific paper.

Objectives

• The student is able to critically evaluate and adapt AI algorithms, visualization techniques, and their integration for interactively accessing complex datasets
• The student is able to design multimedia analytics solutions for contemporary (societal) problems taking account domain specific constraints
• The student is able to develop multimedia analytics solutions
• The student is able to assess scientific multimedia analytics developments, define a multimedia analytics research project, and execute the project by engineering, evaluation, and reporting.
• The student is able to develop interactive visualizations to effectively communicate complex data and AI concepts for a target user group

Teaching methods

• Lecture
• Presentation/symposium
• Supervision/feedback meeting

Learning activities

Attendance

This programme does not have requirements concerning attendance (OER part B).

Additional requirements for this course:

The presentation at the final day of the course and the meeting with your TA are all mandatory. 

Assessment

Assignments

1. Group assignment: intermediate report (20%)

A schematic report introducing the problem you aim to address, the design of your multimedia analytics solution in terms of the interface, the scientific embedding and innovation,  and the software infrastructure in Plotly / Dash for realizing it. The report should be 2-3 pages (excluding references) in the IEEE VIS template. This report, especially the introduction, design and the scientific embedding and innovation are also the starting point for your scientific report.  

A more elaborate description of this deliverable can be found under assignments in Canvas.

2. Group assignment: demo (30%)

Your work should result in a working demo to be presented at the final conference day (June 27, mandatory presence) for your peers and UvA staff members and provided as a software repository for further evaluation. 

A more elaborate description  of this deliverable can be found under assignments in Canvas.

3. Group assignment: scientific report (50%)

This is where you report on your Multimedia Analytics solution in the form of a scientific paper conform the format and guidelines of the IEEE VIS conference with 6-8 pages (so shorter than the regular VIS paper) . In a separate document you should provide a clear description of the individual contributions to the project. 

A more elaborate description of  this deliverable can be under assignments in Canvas..

Fraud and plagiarism

The 'Regulations governing fraud and plagiarism for UvA students' applies to this course. This will be monitored carefully. Upon suspicion of fraud or plagiarism the Examinations Board of the programme will be informed. For the 'Regulations governing fraud and plagiarism for UvA students' see: www.student.uva.nl

Course structure

Contact information

Coordinator

• prof. dr. Marcel Worring

Staff

• G. Barreto Ferreira Marcelino
• F. Gholamzadeh Nasrabadi MSc
• Conor Mc Carthy
• Y. Mohamadi MSc
• I. Najdenkoska MSc
• Y. Zheng MSc