Virtual Hairstyling Try-On (VTO) Methods and Systems Using Generative AI

Abstract

Virtual Hairstyle Try-On (VTO) Methods and Systems Utilizing Generative AI: In some embodiments, computer systems and methods provide a VTO hairstyle experience. A generative artificial intelligence (Gen AI) model (e.g., a diffusion-based model and a conditioning network) configured to generate output images in response to spatial conditioning is called with a plurality of conditioning images to condition the generation of the new hairstyle onto a face image. The conditioning images include a hairstyle mask to control the shape of the new hairstyle and an edge detection result image to guide the structure of the new hairstyle. The hairstyle mask and the edge detection result are obtained from a 3D model of a hairstyle specimen.The mask is aligned with a facial pose, for example, one determined from a generated facial mesh. The face image is preprocessed to remove any existing hairstyle (for example, using a stretched hair mask and a fill). Figure for the abbreviation: none.

Images