Multi-task, Multi-domain Learning: application to semantic segmentation and pose regression - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Neurocomputing Année : 2017

Multi-task, Multi-domain Learning: application to semantic segmentation and pose regression

Rémi Emonet
Elisa Fromont
Damien Muselet
Alain Trémeau
  • Fonction : Auteur
  • PersonId : 859601

Résumé

We present an approach that leverages multiple datasets annotated for different tasks (e.g., classification with different labelsets) to improve the predictive accuracy on each individual dataset. Domain adaptation techniques can correct dataset bias but they are not applicable when the tasks differ, and they need to be complemented to handle multi-task settings. We propose a new selective loss function that can be integrated into deep neural networks to exploit training data coming from multiple datasets annotated for related but possibly different label sets. We show that the gradient-reversal approach for domain adaptation can be used in this setup to additionally handle domain shifts. We also propose an auto-context approach that further captures existing correlations across tasks. Thorough experiments on two types of applications (semantic segmenta-tion and hand pose estimation) show the relevance of our approach in different contexts.
Fichier principal
Vignette du fichier
Multi-task_Multi-domain_Learning.pdf (5.55 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01507132 , version 1 (14-04-2017)
hal-01507132 , version 2 (04-07-2017)

Identifiants

Citer

Damien Fourure, Rémi Emonet, Elisa Fromont, Damien Muselet, Natalia Neverova, et al.. Multi-task, Multi-domain Learning: application to semantic segmentation and pose regression. Neurocomputing, 2017, 251, pp.68-80. ⟨10.1016/j.neucom.2017.04.014⟩. ⟨hal-01507132v2⟩
433 Consultations
930 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More