Mashiro

[Bug] Cannot install mmcls requirements after installing mmdet.

1

Thanks for your error report and we appreciate it a lot. **Checklist** 1. I have searched related issues but cannot get the expected help. 2. I have read the [FAQ...

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand...

[Feature] Add TTA transform

1

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand...

[Refine] refine contributing.md

2

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand...

关于GFocal head中的GN

9

翔哥你好，我重新了试了以下gflv2，的确提点还是很多的，并且我将topk4操作换成了普通卷积，也没有发现明显的掉点，现在已经适合部署了。 但是我发现如果我把默认的GN换成BN，fintune的时候loss和gn没有太大差别，但是map就只有0.0几了，请问这是什么情况？ 我把shared head 改成 private head，发现map又正常了。请问到底是啥情况

Can I substitute the bridge（topk，fc or 1x1conv） betwenn LQE and Regbranch with some simple operators?such as variance or some parameters which can describe DFL's shape

4

Hi! I've been following gfl for some time, and it's a great work. Actually, the uncertainty of bbox's distribution could produce some bad results sometime in gfl1. It seems that...

请问如何reconstruct 口罩的uv texture map

3

请问如何reconstruct 口罩的uv texture map？我有一些自己找的口罩图片极其mask，请问我该如何得到对应的texture？或者我想给人脸增加新的饰品，如眼睛，耳机等，应该如何获取对应物体的mask，以及uv_mask?

Question about Varifocal loss

7

In the paper, the negtive weight of BCE loss is **alpha*p^gamma**. However, in varifocal_loss.py， the loss is implemented by: focal_weight = target * (target > 0.0).float() + \ alpha *...

Although there is args.trunck can be resnet50 or resnet101. But I can't find relevant code in the code. Is this project only use widerresnet?

Have you tried PAA assign as anchor assigner?

1

As shown in **Table 10**(USB: Universal-Scale Object Detection Benchmark), PAA has a better performance than atss. Actually, the difference between paa and atss **is the way of IOU awareness** and...