Grad-SLAM: Explaining Convolutional Autoencoders’ Latent Space of Satellite Image Time Series - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue IEEE Geoscience and Remote Sensing Letters Année : 2023

Grad-SLAM: Explaining Convolutional Autoencoders’ Latent Space of Satellite Image Time Series

Résumé

This paper introduces a tool for explaining the latent space generated by applying convolutional autoencoders to satellite image time series, entitled Grad-SLAM. We rely on backpropagated gradient interpretation combined with network activation localization. We use the proposed formula for multiple layers of the encoder, then scale and merge the results to generate a single date contribution metric for the generation of the latent space. We illustrate the potential of this method with the study of the unsupervised classification of agricultural Sentinel-1 time series. We show that critical characterizing dates for unsupervised retrieval of a given class are conditioned by the crop type's radiometric signature and class count. We also present how Grad-SLAM can be used to enhance the understanding of unsupervised classification confusion.
Fichier principal
Vignette du fichier
_Final__GRSL___Sevilla-3.pdf (792.18 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04215828 , version 1 (22-09-2023)

Licence

Paternité

Identifiants

Citer

Thomas Di Martino, Régis Guinvarc’h, Laetitia Thirion-Lefevre, Élise Colin. Grad-SLAM: Explaining Convolutional Autoencoders’ Latent Space of Satellite Image Time Series. IEEE Geoscience and Remote Sensing Letters, 2023, 20, pp.1-5. ⟨10.1109/LGRS.2023.3302906⟩. ⟨hal-04215828⟩
12 Consultations
73 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More