Authors:
Marc Blanchon
1
;
Olivier Morel
1
;
Yifei Zhang
1
;
Ralph Seulin
1
;
Nathan Crombez
2
and
Désiré Sidibé
1
Affiliations:
1
ImViA EA 7535, ERL VIBOT CNRS 6000, Université de Bourgogne Franche Comté (UBFC), 12 Rue de la Fonderie, 71200, Le Creusot and France
;
2
EPAN Research Group, University of Technology of Belfort-Montbliard (UTBM), 90010, Belfort and France
Keyword(s):
Polarimetry, Deep Learning, Segmentation, Augmentation, Reflective Areas.
Related
Ontology
Subjects/Areas/Topics:
Applications
;
Computer Vision, Visualization and Computer Graphics
;
Features Extraction
;
Image and Video Analysis
;
Pattern Recognition
;
Robotics
;
Segmentation and Grouping
;
Software Engineering
Abstract:
In this paper, we propose a novel method for pixel-wise scene segmentation application using polarimetry. To address the difficulty of detecting highly reflective areas such as water and windows, we use the angle and degree of polarization of these areas, obtained by processing images from a polarimetric camera. A deep learning framework, based on encoder-decoder architecture, is used for the segmentation of regions of interest. Different methods of augmentation have been developed to obtain a sufficient amount of data, while preserving the physical properties of the polarimetric images. Moreover, we introduce a new dataset comprising both RGB and polarimetric images with manual ground truth annotations for seven different classes. Experimental results on this dataset, show that deep learning can benefit from polarimetry and obtain better segmentation results compared to RGB modality. In particular, we obtain an improvement of 38.35% and 22.92% in the accuracy for segmenting windows
and cars respectively.
(More)