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
Hello. I have read your paper and your code carefully. I think this is a very high-quality job.
But when I train the model, I found that some regions are not oval (like the hair in TED, or the eyes in VOX). Here are some examples.
I check the implementation of the flow predictor. It generates the sparse motion by rendering a gaussian heatmap according to the region params.
In my experience, the Gaussian distribution generates elliptical regions.
Considering some regions are not oval, can the sparse motion accurate enough? Or is there some other mechanism to handle this situation? Do you think this could be a direction for improvement in the future?
The text was updated successfully, but these errors were encountered:
Hello. I have read your paper and your code carefully. I think this is a very high-quality job.
![image](https://user-images.githubusercontent.com/11981375/118645260-47978800-b811-11eb-969f-7c35382d0be0.png)
![image](https://user-images.githubusercontent.com/11981375/118645961-2edba200-b812-11eb-94ad-3fc10a50d833.png)
But when I train the model, I found that some regions are not oval (like the hair in TED, or the eyes in VOX). Here are some examples.
I check the implementation of the flow predictor. It generates the sparse motion by rendering a gaussian heatmap according to the region params.
In my experience, the Gaussian distribution generates elliptical regions.
Considering some regions are not oval, can the sparse motion accurate enough? Or is there some other mechanism to handle this situation? Do you think this could be a direction for improvement in the future?
The text was updated successfully, but these errors were encountered: