Furthermore, for simplicity we only use one discriminator in practice, namely, D 1→2 which corresponds to the face-to-cartoon path, our target application.
確定! 回上一頁