You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thank you for your interesting work on the project! I have a question regarding the adaptation of our model to handle additional degradations, such as low-light conditions, blur, and others.
Would it be necessary to modify the prompt pool for this purpose? I’d appreciate your insights on the best approach to take.
The text was updated successfully, but these errors were encountered:
Thanks for your attention! In my view, for more degradations:
You may enlarge prompt pool size to adapt additional degradtions.
We focus on adverse weather restoration, which includes some similarities and difference across different degradations. Thus, such as low-light and noise situation, I consider it will achieve superior performance.
The core motivation of this operation is to utilize unique characteristics and commonalities between degradations. So theoretically, whatever degradations there are, the workflow should be worked. This is because many degradations in the real world often have these properties.
Thank you for your interesting work on the project! I have a question regarding the adaptation of our model to handle additional degradations, such as low-light conditions, blur, and others.
Would it be necessary to modify the prompt pool for this purpose? I’d appreciate your insights on the best approach to take.
The text was updated successfully, but these errors were encountered: