arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.10261cs.CVcs.CRcs.LG
收藏

一种用于图像拷贝检测的自监督描述子

A Self-Supervised Descriptor for Image Copy Detection

Meta AI

Ed Pizzi, Sreya Dutta Roy, Sugosh Nagavara Ravindra, Priya Goyal, Matthijs Douze

摘要:

图像拷贝检测是内容审核的重要任务。我们介绍了SSCD,这是一个建立在最近的自我监督对比训练目标上的模型。我们通过改变架构和训练目标,包括来自实例匹配文献的池化算子,以及将对比学习适应于结合图像的增强,来使这种方法适应复制检测任务。我们的方法依赖于熵正则化项,促进描述符向量之间的一致分离,我们证明,这显着提高了复制检测的准确性。我们的方法产生了一个紧凑的描述符向量,适用于现实世界的网络规模的应用程序。可以将来自背景图像分布的统计信息并入描述符中。在最近的DISC2021基准测试中,在所有设置中,SSCD的性能都超过了基线复制检测模型和为图像分类设计的自监督架构。例如,SSCD的性能比Simplified描述符高出48%。代码可在https://github.com/facebookresearch/sscd-copy-detection上获得。

英文摘要:

Image copy detection is an important task for content moderation. We introduce SSCD, a model that builds on a recent self-supervised contrastive training objective. We adapt this method to the copy detection task by changing the architecture and training objective, including a pooling operator from the instance matching literature, and adapting contrastive learning to augmentations that combine images. Our approach relies on an entropy regularization term, promoting consistent separation between descriptor vectors, and we demonstrate that this significantly improves copy detection accuracy. Our method produces a compact descriptor vector, suitable for real-world web scale applications. Statistical information from a background image distribution can be incorporated into the descriptor. On the recent DISC2021 benchmark, SSCD is shown to outperform both baseline copy detection models and self-supervised architectures designed for image classification by huge margins, in all settings. For example, SSCD out-performs SimCLR descriptors by 48% absolute. Code is available at https://github.com/facebookresearch/sscd-copy-detection.

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。

↑