no code implementations • 11 May 2018 • Feiwu Yu, Xinxiao wu, Yuchao Sun, Lixin Duan
By taking advantage of these two-level adversarial learning, our method is capable of learning a domain-invariant feature representation of source images and target videos.