Polarimetric image augmentation - Informatique, Biologie Intégrative et Systèmes Complexes Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Polarimetric image augmentation

Résumé

This paper deals with new augmentation methods for an unconventional imaging modality sensitive to the physics of the observed scene called polarimetry. In nature, polarized light is obtained by reflection or scattering. Robotics applications in urban environments are subject to many obstacles that can be specular and therefore provide polarized light. These areas are prone to segmentation errors using standard modalities but could be solved using information carried by the polarized light. Deep Convolutional Neural Networks (DCNNs) have shown excellent segmentation results, but require a significant amount of data to achieve best performances. The lack of data is usually overcomed by using augmentation methods. However, unlike RGB images, polarization images are not only scalar (intensity) images and standard augmentation techniques cannot be applied straightforwardly. We propose enhancing deep learning models through a regularized augmentation procedure applied to polarimetric data in order to characterize scenes more effectively under challenging conditions. We subsequently observe an average of 18.1% improvement in IoU between not augmented and regularized training procedures on real world data.
Fichier principal
Vignette du fichier
2020_marc_icpr_1.pdf (11.97 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02977822 , version 1 (26-10-2020)

Identifiants

  • HAL Id : hal-02977822 , version 1

Citer

Marc Blanchon, Olivier Morel, Fabrice Meriaudeau, Ralph Seulin, Désiré Sidibé. Polarimetric image augmentation. 25th International Conference on Pattern Recognition (ICPR 2020), Jan 2021, Milan, Italy. ⟨hal-02977822⟩
152 Consultations
80 Téléchargements

Partager

Gmail Facebook X LinkedIn More