Work

Translation-classification loss for SAR image understanding with deep learning

Year: 2025

Type: article

Abstract: SAR-to-optical translator networks are especially used to overcome the lack of optical images under cloudy conditions. Those translations being used for downstream tasks, they require the reconstruction of reliable patterns with respect to the underlying objects. In this paper, we propose a novel training strategy to account for land-cover complexity through a conjoint Translation-Classification Loss (TCL). The proposed loss evaluates the classifiability of translated images with a pre-trained land-cover classifier by assessing the reliability of its predictions and the relevance of its extracted hidden features. This new loss is applied to nine translators from the literature and to a tenth architecture introduced in the paper. Experiments show that applying the TCL not only improves the credibility of structures, patterns and textures but it also allows for better class discrimination and transitions while avoiding unreliable hallucinated artifacts produced by standard losses in adversarial approaches.

Source: Computer Vision and Image Understanding

DOI: 10.1016/j.cviu.2025.104374

Authors: Antoine Bralet, Abdourrahmane M. Atto, Jocelyn Chanussot, Emmanuel Trouvé

Institutions: Université Savoie Mont Blanc, Institut polytechnique de Grenoble, Centre National de la Recherche Scientifique, Institut national de recherche en sciences et technologies du numérique, Université Grenoble Alpes

Language: English

Topic: Remote-Sensing Image Classification

Subfield: Media Technology

Field: Engineering

Domain: Physical Sciences

Open Access status: hybrid

Estimated APC paid: $2,090

Funder: Région Auvergne-Rhône-Alpes