Physically-admissible polarimetric data augmentation for road-scene analysis. (arXiv:2206.07431v1 [cs.CV])

Web: http://arxiv.org/abs/2206.07431

Polarimetric imaging, along with deep learning, has shown improved
performances on different tasks including scene analysis. However, its
robustness may be questioned because of the small size of the training
datasets. Though the issue could be solved by data augmentation, polarization
modalities are subject to physical feasibility constraints unaddressed by
classical data augmentation techniques. To address this issue, we propose to
use CycleGAN, an image translation technique based on deep generative models
that solely relies on unpaired data, to transfer …

analysis arxiv augmentation cv data