arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2510.17014cs.CV
收藏

卫星任务需要专门的预训练吗?

Do Satellite Tasks Need Special Pretraining?

YerevaNN research lab and YSU(YerevaNN研究实验室和YSU);Saint Louis University(圣路易斯大学);University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)

Ani Vanyan, Alvard Barseghyan, Hakob Tamazyan, Tigran Galstyan, Vahan Huroyan, Naira Hovakimyan, Hrant Khachatrian

AI总结:

本文通过低分辨率遥感泛化基准及在 MillionAID 上训练改进版 iBOT,检验专用遥感预训练的必要性,发现 ViT-B 规模下并无稳定优于通用基线的收益。

摘要:

基础模型推动了机器学习在包括图像在内的多种模态上的发展。最近,多个团队训练了专门面向遥感应用的基础模型。这类研究的动机在于遥感影像具有独特特征、具体应用以及对卫星图像分析有用的特定鲁棒性类型。在本工作中,我们系统性地质疑如下观点:至少在小规模情形下,专用基础模型比通用视觉基础模型更有用。首先,我们设计了一个简单的基准,用于衡量遥感模型在两个下游任务上对较低分辨率图像的泛化能力。其次,我们在 MillionAID(一个 ImageNet 规模的卫星影像数据集)上训练 iBOT 这一自监督视觉编码器,并加入若干针对遥感的修改。我们表明,在 ViT-B 规模下,这些预训练模型都没有相对通用基线带来一致的改进。

英文摘要:

Foundation models have advanced machine learning across various modalities, including images. Recently multiple teams trained foundation models specialized for remote sensing applications. This line of research is motivated by the distinct characteristics of remote sensing imagery, specific applications and types of robustness useful for satellite image analysis. In this work we systematically challenge the idea that specific foundation models are more useful than general-purpose vision foundation models, at least in the small scale. First, we design a simple benchmark that measures generalization of remote sensing models towards images with lower resolution for two downstream tasks. Second, we train iBOT, a self-supervised vision encoder, on MillionAID, an ImageNet-scale satellite imagery dataset, with several modifications specific to remote sensing. We show that none of those pretrained models bring consistent improvements upon general-purpose baselines at the ViT-B scale.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑