Cross-database representation and transfer learning of facial expressions

Our face is a key modality to convey emotions and infer intention. This makes face analysis an important factor in understanding the underlying mechanisms of interaction. Automatic solutions for facial expression recognition promise to deliver a significant fraction of the currently missing componen...

Full description

Bibliographic Details
Main Author: Almaev, Timur
Format: Thesis (University of Nottingham only)
Language:English
Published: 2018
Subjects:
Online Access:https://eprints.nottingham.ac.uk/48033/
_version_ 1848797676106153984
author Almaev, Timur
author_facet Almaev, Timur
author_sort Almaev, Timur
building Nottingham Research Data Repository
collection Online Access
description Our face is a key modality to convey emotions and infer intention. This makes face analysis an important factor in understanding the underlying mechanisms of interaction. Automatic solutions for facial expression recognition promise to deliver a significant fraction of the currently missing component of non-verbal communication to the human-machine interaction enabling more fulfilling experience closely modelling interpersonal communication. This thesis presents three major contributions aimed to overcome a number of issues currently preventing modern face analysis solutions from being applied in practice. The problem of reliable automatic discovery of facial actions is first considered from the point of view of manual feature craft, exploring ways to highlight features related to interpersonal commonalities in facial expression appearance, disregarding those corresponding to environmental conditions and subjective differences. It is then approached from the Multi-Task and Transfer learning perspective, presenting solutions for cost and performance efficient training of facial expression detection algorithms. Finally, a novel solution is proposed for multi-database heterogeneous data representation aimed to provide an environment for better generalisable face analysis solutions training and evaluation.
first_indexed 2025-11-14T20:07:39Z
format Thesis (University of Nottingham only)
id nottingham-48033
institution University of Nottingham Malaysia Campus
institution_category Local University
language English
last_indexed 2025-11-14T20:07:39Z
publishDate 2018
recordtype eprints
repository_type Digital Repository
spelling nottingham-480332025-02-28T13:55:13Z https://eprints.nottingham.ac.uk/48033/ Cross-database representation and transfer learning of facial expressions Almaev, Timur Our face is a key modality to convey emotions and infer intention. This makes face analysis an important factor in understanding the underlying mechanisms of interaction. Automatic solutions for facial expression recognition promise to deliver a significant fraction of the currently missing component of non-verbal communication to the human-machine interaction enabling more fulfilling experience closely modelling interpersonal communication. This thesis presents three major contributions aimed to overcome a number of issues currently preventing modern face analysis solutions from being applied in practice. The problem of reliable automatic discovery of facial actions is first considered from the point of view of manual feature craft, exploring ways to highlight features related to interpersonal commonalities in facial expression appearance, disregarding those corresponding to environmental conditions and subjective differences. It is then approached from the Multi-Task and Transfer learning perspective, presenting solutions for cost and performance efficient training of facial expression detection algorithms. Finally, a novel solution is proposed for multi-database heterogeneous data representation aimed to provide an environment for better generalisable face analysis solutions training and evaluation. 2018-07-19 Thesis (University of Nottingham only) NonPeerReviewed application/pdf en arr https://eprints.nottingham.ac.uk/48033/1/Timur_Almaev_Final_Submission.pdf Almaev, Timur (2018) Cross-database representation and transfer learning of facial expressions. PhD thesis, University of Nottingham. Affective computing facial expression recognition transfer learning multi-task learning machine learning
spellingShingle Affective computing
facial expression recognition
transfer learning
multi-task learning
machine learning
Almaev, Timur
Cross-database representation and transfer learning of facial expressions
title Cross-database representation and transfer learning of facial expressions
title_full Cross-database representation and transfer learning of facial expressions
title_fullStr Cross-database representation and transfer learning of facial expressions
title_full_unstemmed Cross-database representation and transfer learning of facial expressions
title_short Cross-database representation and transfer learning of facial expressions
title_sort cross-database representation and transfer learning of facial expressions
topic Affective computing
facial expression recognition
transfer learning
multi-task learning
machine learning
url https://eprints.nottingham.ac.uk/48033/