本文介绍了一种简单的目标本地化方法（LOST），利用预训练的视觉转换器的激活特征，实验证明该方法在目标发现方面优于其他方法。同时，训练一个不具有类别属性的检测器可以进一步提高性能。此外，该方法在无监督对象发现任务上也有潜力。

本文提出了一种名为LOST的简单方法，用于在图像集合中定位目标，无需昂贵的注释活动。该方法利用自我监督方式预训练的视觉转换器的激活特征，实验结果表明，该方法在PASCAL VOC 2012上的表现优于最先进的目标发现方法，最高可达8 CorLoc点。通过训练一个不具有类别属性的检测器，可以再次提高7个点。在无监督对象发现任务上也展示了有希望的结果。

If you don’t care about the Who, Where, When, Why, How and the Why Is It A Lost Chapter? and want to skip to the What: I wrote a new chapter for Writing An Interpreter In Go and you can read it...

Yesterday I lost all my blog data in a rather funny way. When I installed this new blog engine, that is basically a Lamer News slightly modified to serve as a blog, I spinned a Redis instance...

Everyone at worked has been talking about Lost incessantly since I started working at skinnyCorp. It was driving me mad - mostly because they kept talking about all sorts or interesting things....