Microsoft developed a diffusion model capable of converting a single two-dimensional picture of a human face into a three-dimensional (3D) avatar. The 3D avatar can then be utilised to offer a realistic 3D view of the person for gaming or other purposes, or to create a virtual reality (VR) or augmented reality (AR) experience. The diffusion model is based on a machine-learning method known as the diffusion model (ADM). To train the model, the researchers employed a dataset of 200,000 3D face models. The model was able to construct a realistic metaverse from a single 2D photograph. Studies propose a 3D generative model for the automatic development of 3D digital avatars, but generating the rich characteristics required for high-quality avatars is challenging due to the prohibitive memory and processing needs associated with 3D. The roll-out diffusion network (Rodin) is suggested to overcome this problem.
Where Industry Meet Trends
Menu
English