Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

实验输入数据问题 #13

Closed
AirCL30 opened this issue Jul 12, 2024 · 5 comments
Closed

实验输入数据问题 #13

AirCL30 opened this issue Jul 12, 2024 · 5 comments
Labels
question Further information is requested

Comments

@AirCL30
Copy link

AirCL30 commented Jul 12, 2024

❔Question

Additional context

作者你好,感谢你和你的团队做出的贡献!
我在阅读你的论文和代码时,存在些疑问,烦请你帮忙给出解答。
训练数据在进入到FCNet前经过QTNet的预处理,source images和target images会相应的增加source feak images和target feak images,并且手动将source feak images和target feak images分别放入到原source images文件夹和target images文件夹中,送入到FCNet进行训练
图片
如此做法,source images文件夹中包含了原始的source images和source feak images,这二者的域是不相同的,source feak images应该更偏向于target域,是不是不应该直接作为source来处理。同样的,target images文件夹中也是如此,target feak images已经较为干净,应该更偏向于source域。但是我在代码中没有看到对这些数据进行区分,source images文件夹中的所有图片混为了一类,target images文件夹中的图片也混为了一类。我认为这种做法是没有问题,但是如果仅仅使用原来的source images和从QTNet得到的source feak images训练YOLO,二者都有标签,效果理论上来说应该也是不错的,从这篇工作来看,domain adaptation的意义在于哪里呢?
图片
烦请回复,感谢!

@AirCL30 AirCL30 added the question Further information is requested label Jul 12, 2024
Copy link

👋 Hello @AirCL30, thank you for your interest in YOLOv5 🚀! Please visit our ⭐️ Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution.

If this is a 🐛 Bug Report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom training ❓ Question, please provide as much information as possible, including dataset images, training logs, screenshots, and a public link to online W&B logging if available.

For business inquiries or professional support requests please visit https://ultralytics.com or email Glenn Jocher at glenn.jocher@ultralytics.com.

Requirements

Python>=3.6.0 with all requirements.txt installed including PyTorch>=1.7. To get started:

$ git clone https://github.com/ultralytics/yolov5
$ cd yolov5
$ pip install -r requirements.txt

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

Status

CI CPU testing

If this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training (train.py), validation (val.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

@qinhongda8
Copy link
Owner

1)source fake 和target fake images 的目标的生成中间域以减少源域与目标域之间的间隙,这是领域自适应中的一个常见思路,具体建议可查阅相关文献。2)“如果仅仅使用原来的source images和从QTNet得到的source feak images训练YOLO” 可以看作是补充YOLO+QTNet的消融实验,而且在一些场景下(例如雨天)单纯图像转换的结果并不理想。此外,图像转换也是领域自适应的一种方法。

@AirCL30
Copy link
Author

AirCL30 commented Jul 23, 2024

感谢你的回答!但是你似乎没有回答我的问题。
“source images文件夹中包含了原始的source images和source feak images,这二者的域是不相同的,source feak images应该更偏向于target域,是不是不应该直接作为source来处理。同样的,target images文件夹中也是如此,target feak images已经较为干净,应该更偏向于source域。但是我在代码中没有看到对这些数据进行区分,source images文件夹中的所有图片混为了一类,target images文件夹中的图片也混为了一类。我认为这种做法是没有问题,但是如果仅仅使用原来的source images和从QTNet得到的source feak images训练YOLO,二者都有标签,效果理论上来说应该也是不错的,从这篇工作来看,domain adaptation的意义在于哪里呢?”
我的问题在于,你在代码中把source images和source feak images混为一个source域,同时把target images和target feak images混为target域,我在代码中没有看到对这些数据域进行区分,这种做法是不是不太正确?还是我的理解有误?
你的回答避开了我的问题,通过中间域以减少源域与目标域之间的间隙是领域自适应中的一个常见思路,但是你的代码似乎不太符合这种思路。不知道是不是我的理解问题。麻烦解答。感谢!

@qinhongda8
Copy link
Owner

不需要区分的,例如在Foggy cityscapes实验中,用正常天气图像生成的假雾天图像,是为了让源域中有标注的数据中能拥有接近目标域(雾天)特征的数据。生成各自域假图像的目的是降低在特征对齐中的域间隙,因为域间隙越大对抗学习的效果越差。如果加以区分成4个单独域或者像你说的区分方法,对于后续的对抗学习就没有帮助了(或者你有其他的数据区分思路)。当然这个工作对于一些复杂场景也还存在局限性,

@AirCL30
Copy link
Author

AirCL30 commented Jul 24, 2024

好的,感谢你的解答!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
question Further information is requested
Projects
None yet
Development

No branches or pull requests

2 participants