Unsupervised Domain Adaptation for Personalized Facial Emotion Recognition

Zen, Gloria; Sangineto, Enver; Ricci, Elisa; Sebe, Nicu

doi:10.1145/2663204.2663247

Cited by 45 publications

(66 citation statements)

References 20 publications

(41 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our study, we propose to use transductive parameter transfer [30,35], an adaptive approach which uses transfer learning to overcome this issue by computing a personalised decision surface for each subject based on the similarity of a test subject's data distribution with those of multiple individuals in a training set.…”

Section: The Nature Of Speech and Body Movementsmentioning

confidence: 99%

“…In this paper, we propose to use transfer learning to enable the adaptation of a learnt ensemble model of speaking behaviour to a new unseen subject, based only on unlabelled data. The proposed method, transductive parameter transfer [35], has never been used for this problem. With this method, we provide a solution that can generalise over large populations without requiring personal labelled data.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Personalised models for speech detection from body movements using transductive parameter transfer

Gedik

Hung

2017

Pers Ubiquit Comput

View full text Add to dashboard Cite

We investigate the task of detecting speakers in crowded environments using a single body worn triaxial accelerometer. Detection of such behaviour is very challenging to model as people's body movements during speech vary greatly. Similar to previous studies, by assuming that body movements are indicative of speech, we show experimentally, on a real-world dataset of 3 h including 18 people, that transductive parameter transfer learning (Zen et al. in Proceedings of the 16th international conference on multimodal interaction. ACM, 2014) can better model individual differences in speaking behaviour, significantly improving on the state-of-the-art performance. We also discuss the challenges introduced by the in-thewild nature of our dataset and experimentally show how they affect detection performance. We strengthen the need for an adaptive approach by comparing the speech detection problem to a more traditional activity (i.e. walking). We provide an analysis of the transfer by considering different source sets which provides a deeper investigation of the nature of both speech and body movements, in the context of transfer learning.

show abstract

Section: The Nature Of Speech and Body Movementsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Personalised models for speech detection from body movements using transductive parameter transfer

Gedik

Hung

2017

Pers Ubiquit Comput

View full text Add to dashboard Cite

show abstract

“…A second boosting is applied on the best performing source classifiers to derive the final set of weak classifiers for the target data. In [11], [21], the Adaboost classifiers are replaced with the linear SVMs. First, independent AU classifiers are trained from the source domain data.…”

Section: A Domain Adaptation In Facial Behavior Analysismentioning

confidence: 99%

“…1 More specifically, instead of adjusting the classifier parameters between the domains, as in [10], [21], [22], [23], [11], we propose the use of domainspecific GP experts that model the domain specific attributes. The modeling power of GPs allows us to model the desired attributes in the target domain, in a data-efficient manner.…”

Section: Introductionmentioning

confidence: 99%

Gaussian Process Domain Experts for Modeling of Facial Affect

Eleftheriadis

Rudovic

Deisenroth

et al. 2017

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

Abstract-Most of existing models for facial behavior analysis rely on generic classifiers, which fail to generalize well to previously unseen data. This is because of inherent differences in source (training) and target (test) data, mainly caused by variation in subjects' facial morphology, camera views, etc. All of these account for different contexts in which target and source data are recorded, and thus, may adversely affect the performance of the models learned solely from source data. In this paper, we exploit the notion of domain adaptation and propose a data efficient approach to adapt already learned classifiers to new unseen contexts. Specifically, we build upon the probabilistic framework of Gaussian processes (GPs), and introduce domain-specific GP experts (e.g., for each subject). The model adaptation is facilitated in a probabilistic fashion, by conditioning the target expert on the predictions from multiple source experts. We further exploit the predictive variance of each expert to define an optimal weighting during inference. We evaluate the proposed model on three publicly available datasets for multi-class (MultiPIE) and multi-label (DISFA, FERA2015) facial expression analysis by performing adaptation of two contextual factors: 'where' (view) and 'who' (subject). In our experiments, the proposed approach consistently outperforms (i) both source and target classifiers, while using a small number of target examples during the adaptation, and (ii) related state-ofthe-art approaches for supervised domain adaptation.

show abstract

“…A similar approach, also relying on the weighting of the training examples to minimise the distribution mismatch, was proposed in [3]. A different idea was followed in [20] and in [26], where the authors proposed to learn discriminative mappings from the space of training distributions to the parameter space. To this end, they trained a set of person-specific models, used as the training examples to learn the mapping.…”

Section: Literature Reviewmentioning

confidence: 99%

Learning to Transfer: Transferring Latent Task Structures and Its Application to Person-Specific Facial Action Unit Detection

Almaev

Martínez

Valstar

2015

2015 IEEE International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

In this article we explore the problem of constructing person-specific models for the detection of facial Action Units (AUs), addressing the problem from the point of view of Transfer Learning and Multi-Task Learning. Our starting point is the fact that some expressions, such as smiles, are very easily elicited, annotated, and automatically detected, while others are much harder to elicit and to annotate. We thus consider a novel problem: all AU models for the target subject are to be learnt using person-specific annotated data for a reference AU (AU12 in our case), and no data or little data regarding the target AU. In order to design such a model, we propose a novel Multi-Task Learning and the associated Transfer Learning framework, in which we consider both relations across subjects and AUs. That is to say, we consider a tensor structure among the tasks. Our approach hinges on learning the latent relations among tasks using one single reference AU, and then transferring these latent relations to other AUs. We show that we are able to effectively make use of the annotated data for AU12 when learning other person-specific AU models, even in the absence of data for the target task. Finally, we show the excellent performance of our method when small amounts of annotated data for the target tasks are made available.

show abstract

Unsupervised Domain Adaptation for Personalized Facial Emotion Recognition

Cited by 45 publications

References 20 publications

Personalised models for speech detection from body movements using transductive parameter transfer

Personalised models for speech detection from body movements using transductive parameter transfer

Gaussian Process Domain Experts for Modeling of Facial Affect

Learning to Transfer: Transferring Latent Task Structures and Its Application to Person-Specific Facial Action Unit Detection

Contact Info

Product

Resources

About