Comments (4)
您好!关于合成用的数据我们已经在readme里放出链接了,训练用的数据是以online的方式进行合成的。我们的先验也能解决合成数据带来的domain gap的问题(比如后面的可控调节部分)。
关于“虽然在一些数据看似获得不错的结果”这一点,我们的代码已经开源了,您也可以进行比较,视觉上的效果在RTTS这个数据上是普遍有明显提升的,并非只是“一些数据”;另外“看似不错”这一点当然存在主观成分,您认为DAD效果更好也完全没有关系,不过我们在实验的过程中也发现DAD会在一些case上生成很不和谐的红色伪影,而RIDCP也确实获得了最高的US分数。
另外我们的User Study是以尽可能random的方式进行的(图片的挑选,study过程中显示的顺序等均为随机),我认为我们US指标是可靠可信的,并非是“所谓的US”。后面我们可以整理一下User Study的结果放在github上。
from ridcp_dehazing.
To assess the actual contributions, we can leverage the haze rendering pipeline proposed in the paper and established dehazing networks like AECR-Net and FFA-Net to evaluate the practical impact of the codebook prior. Based on my experience, a well-designed codebook prior can effectively address domain gap issues in low-level tasks. Additionally, there is currently a lack of suitable metrics for objectively evaluating image quality when dehazing real-world images.
from ridcp_dehazing.
@mengxzh15 您好,请问你现在在去雾项目上用的是何种模型,我利用作者该仓合成了数据,并进行了一些训练,但在我真实数据上效果依旧欠佳,故来此寻求帮助,请问您最终话现在用的哪个模型法。作者的在线数据合成模块也是可以作为插件,扦插到其他的去雾代码中。
祝好
tongchangD
from ridcp_dehazing.
@mengxzh15 可能你是对的,我将作者的chm模型什么的都换掉了,直接换成了一个基础的Unet模型,使用相同的数据合成方案,结果是相差无几的。在真实数据上效果都比较差,我现在在尝试使用DAD的方案来合成数据。
from ridcp_dehazing.
Related Issues (20)
- 真实场景的雾天数据合成 HOT 14
- Great work! I would like to ask a simple question HOT 2
- 关于模型测试 HOT 9
- 运行Demo的出错疑问,往有空解惑! HOT 2
- 关于训练模型 HOT 2
- 模型部署问题 HOT 3
- Dataset
- 模型训练问题 HOT 1
- quick demo
- 关于视频去雾
- 关于您公布的500张数据集
- 关于Stage I 的预训练
- 关于CHM和VQGAN训练的开源 HOT 1
- Google Drive links HOT 1
- ITS训练集
- 复现问题 HOT 1
- About Model Efficiency HOT 4
- 关于生成有雾图像 HOT 3
- cannot find the CHM pretrained weight HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from ridcp_dehazing.