When: 1 December 2023 at 9.30 to 15.30

Where:  Drivkraften (BMC:E12018), Forum Medicum, Sölvegatan 19, Lund Sweden

Primary audience researchers (estimate: 40-50 attendants)

• who use audio-/video recordings of behaviour in their research
• in image analysis who may find a new research challenge in automated analysis

To participate is free of charge. Sign up for participation at ai.lu.se/2023-12-01/registration/.

Abstract

Researchers across several faculties draw on video recordings of animal and human behaviour (e.g., birds, humans speaking and gesturing in conversation or interviews) as their primary data source. These data types raise several challenges. We therefore arrange a workshop to focus on methodological aspects of video-based research. The aim is to explore challenges and opportunities in a) coding/annotation, b) analysis, and c) data management of human and animal behavior. We focus on automated analysis methods, using machine learning, and data management using FAIR principles which raise ethical questions.

Preliminary program schedule

9.30 Check-in, coffee

10.00 Welcome and introduction

Keynote

Speaker: Baldvin Gislason-Bern, Axis Communications

Opportunities and challenges in video analytics 
Video analytics has been the next big thing in video surveillance for at least 18 years. Has the time finally come? We will investigate the challenges and opportunities that face companies and society when it comes to video analytics, from a technical, economical and an ethical point of view, and discuss how research using video analytics and the business world could find inspiration from each other.

10:50-11.00 Break

11.00-11.30 Presentations

• Recordings of human behaviour
  Marianne Gullberg, Professor, Linguistics, Lund University
• Recordings of animal behaviour 
  Susanne Åkesson, Professor, Evolutionary ecology, Lund University
• Recordings of human-animal interaction
  Susanne Schötz, Senior lecturer, Logopedics, Lund University

11.30-12.00 Group discussions – Identify research domains and interests related to the above

Lunch

13.00-13.15 Short summary of the morning discussion

13.15-13.45 Current and future solutions

• Automated analysis
  Mikael Nilsson, Senior lecturer, Mathematics, Lund University
• Open tools and FAIR data
  Per Runeson, Professor, Computer Science, Lund University

13.50-14.30 Group discussions

• Identify pain points – what is needed for the next significant step forwards, in six months perspective, and in five years?

14.30-15.00 Reporting in plenum – future work

15.00-15.30 Coffee

Organisation

• Marianne Gullberg, Professor, Linguistics, Lund University
• Susanne Åkesson, Professor, Evolutionary ecology, Lund University
• Susanne Schötz, Slecturer, Logopedics, Phoniatrics and Audiology, Lund University
• Mikael Nilsson, Senior lecturer, Mathematics, Lund University
• Per Runeson, Professor, Computer Science, Lund University
• Jonas Wisbrant, NAC, Lund University