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
@HalfSummer11@kfiraberman
Hello, thx for your excellent work. I have read your published paper, but I have some questions about foot skating.
(1) It is said in the paper that "built upon 1D temporal convolution layers, raw outputs tend to suffer from foot skating artifacts", how to explain it? In my opinion, 1D temporal convolution layers ignore the connection relationship between joints, so it will cause foot skating, Right?
(2) How to understand the sentence in the paper that " extract foot contact labels from the content input, use them to correct the feet positions"? Any more detailed explaination for my reference?
Ericbin
Thx a lot!
The text was updated successfully, but these errors were encountered:
@HalfSummer11 @kfiraberman
Hello, thx for your excellent work. I have read your published paper, but I have some questions about foot skating.
(1) It is said in the paper that "built upon 1D temporal convolution layers, raw outputs tend to suffer from foot skating artifacts", how to explain it? In my opinion, 1D temporal convolution layers ignore the connection relationship between joints, so it will cause foot skating, Right?
(2) How to understand the sentence in the paper that " extract foot contact labels from the content input, use them to correct the feet positions"? Any more detailed explaination for my reference?
Ericbin
Thx a lot!
The text was updated successfully, but these errors were encountered: