arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.06122cs.CVcs.LG
收藏

空间一致性表示学习

Spatially Consistent Representation Learning

Kakao Brain

Byungseok Roh, Wuhyun Shin, Ildoo Kim, Sungwoong Kim

摘要:

自监督学习已被广泛用于从未标记图像中获取可迁移的表示。特别是,最近的对比学习方法在下游图像分类任务上表现出令人印象深刻的性能。虽然这些对比方法主要集中于在语义保持变换下在图像级生成不变的全局表示,但它们容易忽视局部表示的空间一致性,因此在目标检测和实例分割等定位任务的预训练中具有局限性。此外,在现有的对比方法中使用的积极裁剪视图可以最小化单个图像的语义不同区域之间的表示距离。在本文中,我们提出了一个空间一致的表示学习算法(SCRL)的多对象和特定位置的任务。特别是,我们设计了一种新的自我监督的目标,试图产生连贯的空间表示的随机裁剪的局部区域根据几何平移和缩放操作。在使用基准数据集的各种下游定位任务上,所提出的SCRL显示出比图像级监督预训练以及最先进的自监督学习方法显著的性能改进。代码可在https://github.com/kakaobrain/scrl上获得

英文摘要:

Self-supervised learning has been widely used to obtain transferrable representations from unlabeled images. Especially, recent contrastive learning methods have shown impressive performances on downstream image classification tasks. While these contrastive methods mainly focus on generating invariant global representations at the image-level under semantic-preserving transformations, they are prone to overlook spatial consistency of local representations and therefore have a limitation in pretraining for localization tasks such as object detection and instance segmentation. Moreover, aggressively cropped views used in existing contrastive methods can minimize representation distances between the semantically different regions of a single image. In this paper, we propose a spatially consistent representation learning algorithm (SCRL) for multi-object and location-specific tasks. In particular, we devise a novel self-supervised objective that tries to produce coherent spatial representations of a randomly cropped local region according to geometric translations and zooming operations. On various downstream localization tasks with benchmark datasets, the proposed SCRL shows significant performance improvements over the image-level supervised pretraining as well as the state-of-the-art self-supervised learning methods. Code is available at https://github.com/kakaobrain/scrl

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。

↑