Closed hhuyujie closed 4 years ago
We utilize HopeNet to find the row, yaw, and pitch of every testing image. Then, we can use this information to find face pairs that have a similar pose (If the difference between the poses is too large, the transfer result would be weird.). Finally, we can replace some regions of the mask of the source image with target mask.
Thank you for the code sharing. Now, I want to replace some regions of the mask of source image with target mask, could you give me some advice, or share the code of Attribute Transfer in the paper? I remember it's said to use HopeNet, but I don't know how to use the output, that is the angle information of roll, pitch, and yaw.