FitMe: Deep Photorealistic 3D Morphable Model Avatars

S i n g l e i n -t h e -wi l d f a c i a l i ma g e S h a p e a n d Re f l e c t a n c e F i t t i n g S h a p e a n d Re f l e c t a n c e F i t t i n g A c q u i r e d F a c i a l Re f l e c t a n c e P h o t o r e a l i s t i c Re n d e r i n g P h o t o r e a l i s t i c Re n d e r i n g S h a p e Re n d e r i n g Figure 1. FitMe reconstructs relightable shape and reflectance maps for facial avatars, from a single (left) or multiple (right) unconstrained facial images, using a reflectance model and differentiable rendering. The results can be photorealistically rendered in common engines.