Progressive multi-source domain adaptation for personalized facial expression recognition

Zeeshan, Muhammad Osama, Pedersoli, Marco, Lameiras Koerich, Alessandro et Granger, Eric. 2025. « Progressive multi-source domain adaptation for personalized facial expression recognition ». IEEE Transactions on Affective Computing.
(Sous presse)

PDF Granger-E-2025-33104.pdf - Version acceptée Accès restreint à : Administrateur seulement jusqu'au 31 octobre 2026. Licence d'utilisation : Tous les droits réservés aux détenteurs du droit d'auteur. Télécharger (8MB) | Demande de copie

Résumé

Personalized facial expression recognition (FER) involves adapting a machine learning model using samples from labeled sources and unlabeled target domains. Given the challenges of recognizing subtle expressions with considerable interpersonal variability, state-of-the-art unsupervised domain adaptation (UDA) methods focus on the multi-source UDA (MSDA) setting, where each domain corresponds to a specific subject, and improve model accuracy and robustness. However, when adapting to a specific target, the diverse nature of multiple source domains translates to a large shift between source and target data. State-of-the-art MSDA methods for FER address this domain shift by considering all the sources to adapt to the target representations. Nevertheless, adapting to a target subject presents significant challenges due to large distributional differences between source and target domains, often resulting in negative transfer. In addition, integrating all sources simultaneously increases computational costs and causes misalignment with the target. To address these issues, we propose a progressive MSDA approach that gradually introduces information from subjects (source domains) based on their similarity to the target subject. This will ensure that only the most relevant sources from the target are selected, which helps avoid the negative transfer caused by dissimilar sources. During adaptation, the source domains are introduced in a curriculum manner. We first exploit the closest sources to reduce the distribution shift with the target and then move towards the furthest while only considering the most relevant sources based on the predetermined threshold. Furthermore, to mitigate catastrophic forgetting caused by the incremental introduction of source subjects, we implemented a density-based memory mechanism that preserves the most relevant historical source samples for adaptation. Our extensive experiments 1 show the effectiveness of our proposed method on challenging FER datasets: Biovid, UNBC-McMaster, Aff-Wild2, and BAH. Further, performance is evaluated on a cross-dataset setting (UNBC-McMaster → BioVid), showing the importance of gradually adapting to source subjects.

Type de document:	Article publié dans une revue, révisé par les pairs
Professeur:	Professeur Pedersoli, Marco Lameiras Koerich, Alessandro Granger, Éric
Affiliation:	Génie des systèmes, Génie logiciel et des technologies de l'information, Génie des systèmes
Date de dépôt:	03 déc. 2025 19:00
Dernière modification:	08 déc. 2025 20:32
URI:	https://espace2.etsmtl.ca/id/eprint/33104

Actions (Authentification requise)

Dernière vérification avant le dépôt