Get Your Embedding Space in Order: Domain-Adaptive Regression for Forest Monitoring

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Li, Sizhuo, Gominski, Dimitri, Brandt, Martin, Tong, Xiaoye, Ciais, Philippe
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917749664514048
author Li, Sizhuo
Gominski, Dimitri
Brandt, Martin
Tong, Xiaoye
Ciais, Philippe
author_facet Li, Sizhuo
Gominski, Dimitri
Brandt, Martin
Tong, Xiaoye
Ciais, Philippe
contents Image-level regression is an important task in Earth observation, where visual domain and label shifts are a core challenge hampering generalization. However, cross-domain regression within remote sensing data remains understudied due to the absence of suited datasets. We introduce a new dataset with aerial and satellite imagery in five countries with three forest-related regression tasks. To match real-world applicative interests, we compare methods through a restrictive setup where no prior on the target domain is available during training, and models are adapted with limited information during testing. Building on the assumption that ordered relationships generalize better, we propose manifold diffusion for regression as a strong baseline for transduction in low-data regimes. Our comparison highlights the comparative advantages of inductive and transductive methods in cross-domain regression.
format Preprint
id arxiv_https___arxiv_org_abs_2405_00514
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Get Your Embedding Space in Order: Domain-Adaptive Regression for Forest Monitoring
Li, Sizhuo
Gominski, Dimitri
Brandt, Martin
Tong, Xiaoye
Ciais, Philippe
Computer Vision and Pattern Recognition
Image-level regression is an important task in Earth observation, where visual domain and label shifts are a core challenge hampering generalization. However, cross-domain regression within remote sensing data remains understudied due to the absence of suited datasets. We introduce a new dataset with aerial and satellite imagery in five countries with three forest-related regression tasks. To match real-world applicative interests, we compare methods through a restrictive setup where no prior on the target domain is available during training, and models are adapted with limited information during testing. Building on the assumption that ordered relationships generalize better, we propose manifold diffusion for regression as a strong baseline for transduction in low-data regimes. Our comparison highlights the comparative advantages of inductive and transductive methods in cross-domain regression.
title Get Your Embedding Space in Order: Domain-Adaptive Regression for Forest Monitoring
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2405.00514