Comments (3)
i am looking forward to your explanation
from restoreformer.
why not answer this question?
from restoreformer.
Most works for blind face restoration mainly focus on the face areas and pay less attention to the background areas. This work is the same and the reason that may affect the restoration of the background is that while restoring, we will match the priors of all the areas of the face image, including the areas of background, from HQ Dictionary, which is learned from high-quality face images. However, different from the face areas that have strong commonalities between different faces, the background of different images is diverse. It is hard to attain highly similar priors for the background of a specific face image. Besides, as described in the paper, we found that the skip connection in the MHCA happens between the attended feature and the priors rather than the degraded feature can attain better restored results in the face areas, which may loss the information of the original background to some extend.
from restoreformer.
Related Issues (20)
- About FID on test datasets HOT 1
- IDD evaluation script HOT 1
- 测试 HOT 1
- How does this compare to CodeFormer? HOT 3
- Use RestoreFormer with custom images HOT 1
- DICTIONARY MODEL
- Training settings in the paper differ from the code HOT 1
- HQDictionary model HOT 1
- Train dataset HOT 1
- When use the HQ dictionary for MHCA. HOT 1
- Could you add a licence?
- HQ_Dictionary model and RestoreFormer model use different number of attention heads HOT 1
- Recommend for large scale training dataset
- How long has the RestoreFormer model been trained? How to control model training time and save model?
- ONNX Conversion
- Training time of the RestoreFormer model
- 关于测试和训练 HOT 1
- 配置文件 HOT 1
- Can I train RestoreFormer on 1024x1024? HOT 2
- Pretrained Models Link not working 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 restoreformer.