Ruopeng Gao

图像生成器（Generator）中的数据分布信息从何而来？

2

您好，最近在拜读你们发表在CVPR 2021的文章 *Data-Free Knowledge Distillation For Image Super-Resolution*，在其中对于图像生成器的论述有一段是`where x is the training sample and p_x(x) is the distribution of the original dataset.`，但是我好像并没有在文中看到这里原始数据集的分布 p_x(x) 是从哪里获得或者计算或者设定的，也好像在你们的仓库中并没有看到对应这篇文章的源码，也可能是我学艺不精看漏了。 因此想请问一下你们，这部分的原始数据集分布p_x(x)是从哪里得到的呢？

Paper: MeMOTR: Long-Term Memory-Augmented Transformer for Multi-Object Tracking Paper Link: https://arxiv.org/abs/2307.15700 Code Link: https://github.com/MCG-NJU/MeMOTR Area: Tracking / Multi-Object Tracking (MOT) Thanks a lot.

About the confused relationship between DetA and predict IDs.

2

In my understanding, if the same detector is used (that is, the position result of the boxes is fixed), no matter how I change the ID assignment of the targets,...

I encountered something strange while visualizing the dataset. As shown below, seq `012`, `0110.jpg`:  *The text upon the bbox represents the identity and visibility.* The `65` and `38` identities...