Hi, thanks for you work. Something about the input of appearance encoder makes me confused. Since the appearance takes as input the reference image which should be the target image of generator (as written in func 'read_single_appearance_input' in pretrain_apperance.py). Would you mind giving a clear interpretation about the input ?
Hi, thanks for you work. Something about the input of appearance encoder makes me confused. Since the appearance takes as input the reference image which should be the target image of generator (as written in func 'read_single_appearance_input' in pretrain_apperance.py). Would you mind giving a clear interpretation about the input ?