arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2106.09259cs.CVcs.AI
收藏

随机CNN看见物体:CNN的一个归纳偏差及其应用

A Random CNN Sees Objects: One Inductive Bias of CNN and Its Applications

Yun-Hao Cao, Jianxin Wu

摘要:

本文首先揭示了一个令人惊讶的发现:无需任何学习,随机初始化的CNN可以出奇地好地定位对象。也就是说,一个CNN有一个归纳的偏见,自然集中在物体上,命名为托拜厄斯(“该对象是在视线”)。这种经验归纳偏差进一步分析,并成功地应用到自监督学习(SSL)。鼓励CNN通过将每个图像转换为具有不同背景的各种版本来学习关注前景对象的表示,其中前景和背景分离由Tobias指导。实验结果表明,所提出的Tobias算法显著改善了下游任务,尤其是目标检测。本文还表明,Tobias在不同大小的训练集上有一致的改进,并且对图像增强的变化更具弹性。代码可在https://github.com/CupidJay/Tobias上获得。

英文摘要:

This paper starts by revealing a surprising finding: without any learning, a randomly initialized CNN can localize objects surprisingly well. That is, a CNN has an inductive bias to naturally focus on objects, named as Tobias ("The object is at sight") in this paper. This empirical inductive bias is further analyzed and successfully applied to self-supervised learning (SSL). A CNN is encouraged to learn representations that focus on the foreground object, by transforming every image into various versions with different backgrounds, where the foreground and background separation is guided by Tobias. Experimental results show that the proposed Tobias significantly improves downstream tasks, especially for object detection. This paper also shows that Tobias has consistent improvements on training sets of different sizes, and is more resilient to changes in image augmentations. Code is available at https://github.com/CupidJay/Tobias.

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。

↑