Yunxuan Zhang, Siwei Zhang, Yue He +3
The paper introduces a one-shot face reenactment framework that can transfer pose and expression using only a single target image by disentangling appearance and shape in latent sp…