Smart image enhancements are disclosed, including: obtaining a representation of a user's face associated with a set of images associated with the user's face; obtaining a set of extrinsic information corresponding to an image of the set of images; determining a modified smoothing map by modifying a model smoothing map to correspond to the representation of the user's face; and determining an enhanced image based at least in part on the set of extrinsic information corresponding to the image, the modified model smoothing map, and the image.