A Mixture of Personalized Experts for Human Affect Estimation
Author(s)Feffer, Michael; Rudovic, Ognjen; Picard, Rosalind W.
MetadataShow full item record
We investigate the personalization of deep convolutional neural networks for facial expression analysis from still images. While prior work has focused on population-based (“one-size-fits-all”) approaches, we formulate and construct personalized models via a mixture of experts and supervised domain adaptation approach, showing that it improves greatly upon non-personalized models. Our experiments demonstrate the ability of the model personalization to quickly and effectively adapt to limited amounts of target data. We also provide a novel training methodology and architecture for creating personalized machine learning models for more effective analysis of emotion state.
DepartmentMassachusetts Institute of Technology. Media Laboratory
Lecture Notes in Computer Science
Springer International Publishing
Feffer, Michael et al. "A Mixture of Personalized Experts for Human Affect Estimation." MLDM 2018: Machine Learning and Data Mining in Pattern Recognition, Lecture Notes in Computer Science, 10935, Springer International Publishing, 2018, 316-330. © 2018 Springer International Publishing AG
Author's final manuscript