Morphable Diffusion: 3D-Consistent Diffusion for Single-image Avatar Creation

4 https://xiyichen.github.io/morphablediffusion/ Figure 1. Morphable diffusion. We introduce a morphable diffusion model to enable consistent controllable novel view synthesis of humans from a single image. Given a single input image (a) and a morphable mesh model with a target facial expression (b) our method directly generates 3D consistent and photo-realistic images from novel viewpoints (c). Using the generated multi-view consistent images, we can reconstruct a coarse 3D model (d) using off-the-shelf neural surface reconstruction methods such as [80].