The paper shows that you swap the landmarks with source shape and target expression and pose. But I cannot find any code to do this. The code seems to use the target landmarks directly. Am I right? I did not read the code thoroughly, so I am not sure about this. If it has, please show me specific place of the code.
Hi,
Thank you for sharing the code.
The paper shows that you swap the landmarks with source shape and target expression and pose. But I cannot find any code to do this. The code seems to use the target landmarks directly. Am I right? I did not read the code thoroughly, so I am not sure about this. If it has, please show me specific place of the code.
Thanks