Lightweight, Pre-trained Transformers for Remote Sensing Timeseries

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tseng, Gabriel, Cartuyvels, Ruben, Zvonkov, Ivan, Purohit, Mirali, Rolnick, David, Kerner, Hannah
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909091636445184
author Tseng, Gabriel
Cartuyvels, Ruben
Zvonkov, Ivan
Purohit, Mirali
Rolnick, David
Kerner, Hannah
author_facet Tseng, Gabriel
Cartuyvels, Ruben
Zvonkov, Ivan
Purohit, Mirali
Rolnick, David
Kerner, Hannah
contents Machine learning methods for satellite data have a range of societally relevant applications, but labels used to train models can be difficult or impossible to acquire. Self-supervision is a natural solution in settings with limited labeled data, but current self-supervised models for satellite data fail to take advantage of the characteristics of that data, including the temporal dimension (which is critical for many applications, such as monitoring crop growth) and availability of data from many complementary sensors (which can significantly improve a model's predictive performance). We present Presto (the Pretrained Remote Sensing Transformer), a model pre-trained on remote sensing pixel-timeseries data. By designing Presto specifically for remote sensing data, we can create a significantly smaller but performant model. Presto excels at a wide variety of globally distributed remote sensing tasks and performs competitively with much larger models while requiring far less compute. Presto can be used for transfer learning or as a feature extractor for simple models, enabling efficient deployment at scale.
format Preprint
id arxiv_https___arxiv_org_abs_2304_14065
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Lightweight, Pre-trained Transformers for Remote Sensing Timeseries
Tseng, Gabriel
Cartuyvels, Ruben
Zvonkov, Ivan
Purohit, Mirali
Rolnick, David
Kerner, Hannah
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine learning methods for satellite data have a range of societally relevant applications, but labels used to train models can be difficult or impossible to acquire. Self-supervision is a natural solution in settings with limited labeled data, but current self-supervised models for satellite data fail to take advantage of the characteristics of that data, including the temporal dimension (which is critical for many applications, such as monitoring crop growth) and availability of data from many complementary sensors (which can significantly improve a model's predictive performance). We present Presto (the Pretrained Remote Sensing Transformer), a model pre-trained on remote sensing pixel-timeseries data. By designing Presto specifically for remote sensing data, we can create a significantly smaller but performant model. Presto excels at a wide variety of globally distributed remote sensing tasks and performs competitively with much larger models while requiring far less compute. Presto can be used for transfer learning or as a feature extractor for simple models, enabling efficient deployment at scale.
title Lightweight, Pre-trained Transformers for Remote Sensing Timeseries
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2304.14065