arXiv 2309.155352023-09-27cs.CVcs.AIcs.LG

从 LAION-5B 到 LAION-EO：使用锚点数据集过滤数十亿图像以提取卫星图像

From LAION-5B to LAION-EO: Filtering Billions of Images Using Anchor Datasets for Satellite Image Extraction

European Space Agency（欧洲空间局）

Mikolaj Czerkawski, Alistair Francis

备注：: Accepted at the ICCV 2023 Workshop "Towards the Next Generation of Computer Vision Datasets: DataComp Track"
发表信息：: ICCV 2023 Workshop "Towards the Next Generation of Computer Vision Datasets: DataComp Track"

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对大型图像语料库特定领域子集提取难的问题，提出基于锚点数据集结合过滤的提取方法，在卫星图像领域验证后发布了高分辨率图文对数据集 LAION-EO，并介绍其获取流程与特性。

摘要：

LAION-5B 等大型数据集包含网络共享的各类图像分布，但从大型图像语料库中提取特定领域子集颇具挑战。本文提出一种基于锚点数据集、结合进一步过滤的提取方法，并在卫星图像领域进行了验证。该方法最终产出了 LAION-EO 数据集——一个源自网络、包含高（像素级）分辨率卫星图像与文本对的数据集。本文还概述了该数据集的获取流程及部分特性。

英文摘要：

Large datasets, such as LAION-5B, contain a diverse distribution of images shared online. However, extraction of domain-specific subsets of large image corpora is challenging. The extraction approach based on an anchor dataset, combined with further filtering, is proposed here and demonstrated for the domain of satellite imagery. This results in the release of LAION-EO, a dataset sourced from the web containing pairs of text and satellite images in high (pixel-wise) resolution. The paper outlines the acquisition procedure as well as some of the features of the dataset.

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。