P2D: a self-supervised method for depth estimation from polarimetry - Université d'Évry Access content directly
Conference Papers Year : 2021

P2D: a self-supervised method for depth estimation from polarimetry


Monocular depth estimation is a recurring subject in the field of computer vision. Its ability to describe scenes via a depth map while reducing the constraints related to the formulation of perspective geometry tends to favor its use. However, despite the constant improvement of algorithms, most methods exploit only colorimetric information. Consequently, robustness to events to which the modality is not sensitive to, like specularity or transparency, is neglected. In response to this phenomenon, we propose using polarimetry as an input for a self-supervised monodepth network. Therefore, we propose exploiting polarization cues to encourage accurate reconstruction of scenes. Furthermore, we include a term of polarimetric regularization to state-of-the-art method to take specific advantage of the data. Our method is evaluated both qualitatively and quantitatively demonstrating that the contribution of this new information as well as an enhanced loss function improves depth estimation results, especially for specular areas.
Fichier principal
Vignette du fichier
2020_marc_icpr_2.pdf (3.01 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-02977824 , version 1 (26-10-2020)


  • HAL Id : hal-02977824 , version 1


Marc Blanchon, Désiré Sidibé, Olivier Morel, Ralph Seulin, Daniel Braun, et al.. P2D: a self-supervised method for depth estimation from polarimetry. 25th International Conference on Pattern Recognition (ICPR 2020), Jan 2021, Milan, Italy. ⟨hal-02977824⟩
113 View
89 Download


Gmail Facebook X LinkedIn More