arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2301.03992cs.CVcs.LGcs.MM
收藏

视觉Transformer是很好的口罩自动贴标机

Vision Transformers Are Good Mask Auto-Labelers

NVIDIA(英伟达);Meta AI, FAIR(Meta人工智能实验室(基础人工智能研究团队));Fudan University(复旦大学);Caltech(加州理工学院)

Shiyi Lan, Xitong Yang, Zhiding Yu, Zuxuan Wu, Jose M. Alvarez, Anima Anandkumar

摘要:

我们提出了Mask Auto-Labeler(MAL),这是一个高质量的基于transformer的掩码自动标记框架,用于仅使用框注释的实例分割。MAL以框裁剪图像为输入,有条件地生成它们的掩码伪标签。我们证明了视觉Transformer是很好的掩码自动标记器。我们的方法显着减少了自动标记和人工注释之间的差距,关于掩模质量。使用MAL生成的掩码训练的实例分割模型几乎可以与完全监督模型的性能相匹配,保留了高达97.4%的完全监督模型的性能。最好的模型在COCO实例分割上实现了44.1\% mAP(test-dev 2017),显著优于最先进的框监督方法。定性结果表明,MAL产生的面具,在某些情况下,甚至比人类的注释。

英文摘要:

We propose Mask Auto-Labeler (MAL), a high-quality Transformer-based mask auto-labeling framework for instance segmentation using only box annotations. MAL takes box-cropped images as inputs and conditionally generates their mask pseudo-labels.We show that Vision Transformers are good mask auto-labelers. Our method significantly reduces the gap between auto-labeling and human annotation regarding mask quality. Instance segmentation models trained using the MAL-generated masks can nearly match the performance of their fully-supervised counterparts, retaining up to 97.4\% performance of fully supervised models. The best model achieves 44.1\% mAP on COCO instance segmentation (test-dev 2017), outperforming state-of-the-art box-supervised methods by significant margins. Qualitative results indicate that masks produced by MAL are, in some cases, even better than human annotations.

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。

↑