arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.01747cs.CV
收藏

Met数据集:艺术品的实例级识别

The Met Dataset: Instance-level Recognition for Artworks

Czech Technical University in Prague(布拉格捷克理工大学);Osaka University(大阪大学);Columbia University(哥伦比亚大学);University of Amsterdam(阿姆斯特丹大学)

Nikolaos-Antonios Ypsilantis, Noa Garcia, Guangxing Han, Sarah Ibrahimi, Nanne Van Noord, Giorgos Tolias

摘要:

这项工作介绍了一个数据集的大规模实例级识别领域的艺术品。所提出的基准表现出许多不同的挑战,如大类间相似性,长尾分布,和许多类。我们依靠大都会博物馆的开放获取收藏来形成一个大约224k个类的大型训练集,其中每个类对应于一个博物馆展览,其中照片是在摄影棚条件下拍摄的。测试主要是在博物馆游客拍摄的描绘展品的照片上进行的,这在训练和测试之间引入了分布偏移。另外,对一组与Met展品无关的图像进行测试,使任务类似于分布外检测问题。建议的基准遵循其他最近的数据集的范例在不同领域的实例级识别,以鼓励领域独立的方法的研究。一些合适的方法进行评估,为未来的比较提供一个测试平台。将自监督和监督对比学习有效地结合起来,训练用于非参数分类的骨干,是一个很有前途的方向。数据集网页:www.example.com

英文摘要:

This work introduces a dataset for large-scale instance-level recognition in the domain of artworks. The proposed benchmark exhibits a number of different challenges such as large inter-class similarity, long tail distribution, and many classes. We rely on the open access collection of The Met museum to form a large training set of about 224k classes, where each class corresponds to a museum exhibit with photos taken under studio conditions. Testing is primarily performed on photos taken by museum guests depicting exhibits, which introduces a distribution shift between training and testing. Testing is additionally performed on a set of images not related to Met exhibits making the task resemble an out-of-distribution detection problem. The proposed benchmark follows the paradigm of other recent datasets for instance-level recognition on different domains to encourage research on domain independent approaches. A number of suitable approaches are evaluated to offer a testbed for future comparisons. Self-supervised and supervised contrastive learning are effectively combined to train the backbone which is used for non-parametric classification that is shown as a promising direction. Dataset webpage: http://cmp.felk.cvut.cz/met/

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。

↑