Extending global-local view alignment for self-supervised learning with remote sensing imagery

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wanyan, Xinye, Seneviratne, Sachith, Shen, Shuchang, Kirley, Michael
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911850734551040
author Wanyan, Xinye
Seneviratne, Sachith
Shen, Shuchang
Kirley, Michael
author_facet Wanyan, Xinye
Seneviratne, Sachith
Shen, Shuchang
Kirley, Michael
contents Since large number of high-quality remote sensing images are readily accessible, exploiting the corpus of images with less manual annotation draws increasing attention. Self-supervised models acquire general feature representations by formulating a pretext task that generates pseudo-labels for massive unlabeled data to provide supervision for training. While prior studies have explored multiple self-supervised learning techniques in remote sensing domain, pretext tasks based on local-global view alignment remain underexplored, despite achieving state-of-the-art results on natural imagery. Inspired by DINO, which employs an effective representation learning structure with knowledge distillation based on global-local view alignment, we formulate two pretext tasks for self-supervised learning on remote sensing imagery (SSLRS). Using these tasks, we explore the effectiveness of positive temporal contrast as well as multi-sized views on SSLRS. We extend DINO and propose DINO-MC which uses local views of various sized crops instead of a single fixed size in order to alleviate the limited variation in object size observed in remote sensing imagery. Our experiments demonstrate that even when pre-trained on only 10% of the dataset, DINO-MC performs on par or better than existing state-of-the-art SSLRS methods on multiple remote sensing tasks, while using less computational resources. All codes, models, and results are released at https://github.com/WennyXY/DINO-MC.
format Preprint
id arxiv_https___arxiv_org_abs_2303_06670
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Extending global-local view alignment for self-supervised learning with remote sensing imagery
Wanyan, Xinye
Seneviratne, Sachith
Shen, Shuchang
Kirley, Michael
Computer Vision and Pattern Recognition
Since large number of high-quality remote sensing images are readily accessible, exploiting the corpus of images with less manual annotation draws increasing attention. Self-supervised models acquire general feature representations by formulating a pretext task that generates pseudo-labels for massive unlabeled data to provide supervision for training. While prior studies have explored multiple self-supervised learning techniques in remote sensing domain, pretext tasks based on local-global view alignment remain underexplored, despite achieving state-of-the-art results on natural imagery. Inspired by DINO, which employs an effective representation learning structure with knowledge distillation based on global-local view alignment, we formulate two pretext tasks for self-supervised learning on remote sensing imagery (SSLRS). Using these tasks, we explore the effectiveness of positive temporal contrast as well as multi-sized views on SSLRS. We extend DINO and propose DINO-MC which uses local views of various sized crops instead of a single fixed size in order to alleviate the limited variation in object size observed in remote sensing imagery. Our experiments demonstrate that even when pre-trained on only 10% of the dataset, DINO-MC performs on par or better than existing state-of-the-art SSLRS methods on multiple remote sensing tasks, while using less computational resources. All codes, models, and results are released at https://github.com/WennyXY/DINO-MC.
title Extending global-local view alignment for self-supervised learning with remote sensing imagery
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2303.06670