Ryan Elson
Thesis Title: Multi-view Representations for Pose Invariant Face Recognition in Man and Machine
Humans have an incredible ability to identify familiar individuals and process speech and expressions despite substantial changes in the visual picture. These can include changes in lighting, viewing angle, distance, occluders and peripheral elements such as glasses and hair, and both static and dynamic shape change of the face due to speech and expression. Despite this highly functional human capability, facial appearance variations pose a significant challenge to computer vision systems. Achieving Pose-Invariant Face Recognition (PIFR) in computer vision systems remains a significant stumbling block to realizing the full potential of face recognition as a passive biometric technology, despite extensive efforts being made. PIFR is achieved effortlessly by the human visual system but at present we do not understand the human system well enough to provide and implement plausible solutions to the clear technological challenges. The aim of this project is to enhance our understanding of how human observers achieve pose invariant recognition of faces in order to inform AI strategies. We will particularly focus on multi-view or pose-aware strategies and compare these against object-based models or pose-agnostic approaches. We will conduct experiments with human participants, including behavioural studies and incorporating neuroimaging methods such as functional magnetic resonance imaging (fMRI) to help understand the brain networks involved in pose invariance in humans. These experiments will include determining whether, and how, human observers utilise the variations in the visual input whilst achieving pose invariant face recognition. We will also conduct computational experiments exploring computer models of pose-invariant dynamic face recognition.
Biography
I took mathematics with statistics, chemistry, biology and psychology at A-level, but at the time chose psychology as a fun 4th option, not expecting to go anywhere with it. However, I thoroughly enjoyed psychology and found studying human behaviour fascinating, so I decided to pursue it at university.
Partway into my BSc in psychology at the University of York I discovered their integrated masters (MSci) course in psychology that focused on neuroscience and neuroimaging, which I promptly swapped onto. I found research into the organisation of the human brain fascinating, particularly the visual brain. Of the visual brain, face and object processing was especially interesting to me, so my master's dissertation investigated the role of a region called the posterior superior temporal sulcus on facial expression processing.
I then moved to the University of Nottingham in 2019 to begin my PhD. My PhD focuses on how we process faces across changes in viewpoint, and includes behavioural experiments, fMRI experiments and computational modelling. The work primarily looks at processing facial motion (e.g., speech) across changes in view but also includes other work such as how the brain processes caricatured faces.
Publications
Sliwinska, M. W., Elson, R., & Pitcher, D. (2020). Dual-site TMS demonstrates causal functional connectivity between the left and right posterior temporal sulci during facial expression recognition. Brain Stimulation.
Johnston, A., Brown, B. B., & Elson, R. (2021) Synchronous facial action binds dynamic facial features. Scientific Reports 11 (7191)
Elson, R., Valstar, M., Schluppeck, D., & Johnston, A. (2021, July). Viewing face space from a different angle. In I-PERCEPTION (Vol. 12, No. 4, pp. 7-8). 1 OLIVERS YARD, 55 CITY ROAD, LONDON EC1Y 1SP, ENGLAND: SAGE PUBLICATIONS LTD.
Sliwinska, M. W., Elson, R., & Pitcher, D. (2021). Stimulating parietal regions of the multiple-demand cortex impairs novel vocabulary learning. Neuropsychologia, 108047.
Elson, R., Schluppeck, D., Valstar, M., & Johnston, A. (2022, April). Taking face space to the extreme: assessing the effect of hyper-caricaturing faces on the fMRI response in the FFA. In I-PERCEPTION (Vol. 51, No. 5, pp. 354-364). 1 OLIVERS YARD, 55 CITY ROAD, LONDON EC1Y 1SP, ENGLAND: SAGE PUBLICATIONS LTD.
Memberships
Applied Vision Association, Experimental Psychological Society.
Digital Society
University of Nottingham
2019 Cohort, +4
Supervisory Team
Prof Alan Johnston
Dr Denis Schluppeck