Frontiers in Animal Biometrics

12 June 2020, 4.00 PM - 12 June 2020, 5.00 PM

Dr Tilo Burghardt, Department of Computer Science, Visual Information Laboratory, SPHERE, Intelligent Systems Laboratory, University of Bristol

Via Zoom, joining instructions below

Abstract

Animal biometrics develops quantified approaches for detecting aspects of the phenotype; that is recognising animal species, individuals, behaviours, or morphological traits. This talk will look at examples of how this field can benefit from recent advances in computer vision and robotics. In particular, three brief case studies will be presented.

First, we will look at how self-attention can be used in deep detection networks to deal with heavy occlusions in camera trap footage of great apes. Secondly, we will look at traditional techniques for encoding and recognising 1D biometric entities as exemplified on great white shark fins. Finally, we will discuss how robotics, vision, and deep learning can be integrated to perform autonomous identification of Friesian cattle from the air. The talk will conclude that the advancement of animal biometrics will require further integration of methodologies among the engineering and scientific disciplines involved.

Biography

Tilo Burghardt is a Senior Lecturer at the Dept of Computer Science, University of Bristol (UK). He graduated with Distinction in Media Computing (Bakk. Medien-Inf.) at Dresden University of Technology (Germany). Subsequently, he received an MSc in Advanced Computing and PhD in Computer Vision from the University of Bristol.

His research focuses on applied computer vision and animal biometrics. He co-authored more than 50 scientific papers in these areas and contributed to establishing animal biometrics as an emerging cross-discipline routed in pattern recognition and computer vision. He is a member of the British Machine Vision Association, the German Academic Foundation, and a Fellow of the Higher Education Academy. He is Associate Editor of IET Computer Vision and acted as chair for the 24th British Machine Vision Conference. Tilo's enthusiasm for computer science and vision is reflected in his dedication to teaching the subject. In 2018 he received the University of Bristol 'Award for Education' for his educational contributions to the Engineering Faculty

http://www.bristol.ac.uk/engineering/people/tilo-burghardt/index.html

Contact information

For any queries, please contact bvi-enquiries@bristol.ac.uk.

Zoom joining instructions: https://us02web.zoom.us/j/82844496679?pwd=NU9MMWRpd0N5NWdzbjFoTEd0bmhjZz09

Meeting ID: 828 4449 6679
Password: 821028

Tilo Burghardt, BVI Seminar speaker

Edit this page