paper-with-me

홈 › Papers

Self-Supervised Pretraining on Satellite Imagery: a Case Study on Label-Efficient Vehicle Detection

2022-10-21 · Jules BOURCIER, Thomas Floquet, Gohar Dashyan, Tugdual Ceillier, Karteek Alahari, Jocelyn Chanussot

In defense-related remote sensing applications, such as vehicle detection on satellite imagery, supervised learning requires a huge number of labeled examples to reach operational performances. Such data are challenging to obtain as it requires military experts, and some observables are intrinsically rare. This limited labeling capability, as well as the large number of unlabeled images available due to the growing number of sensors, make object detection on remote sensing imagery highly relevant for self-supervised learning. We study in-domain self-supervised representation learning for object detection on very high resolution optical satellite imagery, that is yet poorly explored. For the first time to our knowledge, we study the problem of label efficiency on this task. We use the large land use classification dataset Functional Map of the World to pretrain representations with an extension of the Momentum Contrast framework. We then investigate this model's transferability on a real-world task of fine-grained vehicle detection and classification on Preligens proprietary data, which is designed to be representative of an operational use case of strategic site surveillance. We show that our in-domain self-supervised learning model is competitive with ImageNet pretraining, and outperforms it in the low-label regime.

📄 PDF Abstract BibTeX arXiv:2210.11815

Code (0)

등록된 구현이 없습니다.

Tasks

object-detectionObject DetectionRepresentation LearningSelf-Supervised Learningvehicle detection

Similar Papers 제목 키워드 기반

Cross-Scale Pretraining: Enhancing Self-Supervised Learning for Low-Resolution Satellite Imagery for Semantic Segmentation

2026-01-19 · John Waithaka, Gustave Bwirayesu, Moise Busogi arxiv

Self-supervised pretraining in remote sensing is mostly done using mid-spatial resolution (MR) image datasets due to their high availability. Given the release of high-resolution (HR) datasets, we ask how HR datasets can…

Self-Supervised LearningRepresentation LearningSemantic Segmentation

A Tutorial on ALOS2 SAR Utilization: Dataset Preparation, Self-Supervised Pretraining, and Semantic Segmentation

2026-03-16 · Nevrez Imamoglu, Ali Caglayan, Toru Kouyama arxiv

Masked auto-encoders (MAE) and related approaches have shown promise for satellite imagery, but their application to synthetic aperture radar (SAR) remains limited due to challenges in semantic labeling and high noise le…

Semantic Segmentation

SatDINO: A Deep Dive into Self-Supervised Pretraining for Remote Sensing

2025-08-29 · Jakub Straka, Ivan Gruber arxiv

Self-supervised learning has emerged as a powerful tool for remote sensing, where large amounts of unlabeled data are available. In this work, we investigate the use of DINO, a contrastive self-supervised method, for pre…

Self-Supervised LearningRepresentation Learning

Domain Adaptable Self-supervised Representation Learning on Remote Sensing Satellite Imagery

2023-04-19 · Muskaan Chopra, Prakash Chandra Chhipa, Gopal Mengi, Varun Gupta 외

This work presents a novel domain adaption paradigm for studying contrastive self-supervised representation learning and knowledge transfer using remote sensing satellite data. Major state-of-the-art remote sensing visua…

Contrastive LearningDomain AdaptationRepresentation LearningTransfer Learning

Evaluating the Label Efficiency of Contrastive Self-Supervised Learning for Multi-Resolution Satellite Imagery

2022-10-13 · Jules BOURCIER, Gohar Dashyan, Jocelyn Chanussot, Karteek Alahari

The application of deep neural networks to remote sensing imagery is often constrained by the lack of ground-truth annotations. Adressing this issue requires models that generalize efficiently from limited amounts of lab…

Earth ObservationRepresentation LearningSelf-Supervised Learning