SplattingAvatar: Realistic Real-Time Human Avatars with Mesh-Embedded Gaussian Splatting
This addresses the need for efficient and high-quality virtual humans in real-time graphics, such as for gaming or VR, by offering a novel hybrid approach that is compatible with various animation techniques.
The paper tackles the problem of creating photorealistic human avatars for real-time applications by introducing SplattingAvatar, a hybrid 3D representation that combines mesh geometry with Gaussian splatting, achieving rendering speeds of over 300 FPS on a GPU and 30 FPS on mobile devices with state-of-the-art quality.
We present SplattingAvatar, a hybrid 3D representation of photorealistic human avatars with Gaussian Splatting embedded on a triangle mesh, which renders over 300 FPS on a modern GPU and 30 FPS on a mobile device. We disentangle the motion and appearance of a virtual human with explicit mesh geometry and implicit appearance modeling with Gaussian Splatting. The Gaussians are defined by barycentric coordinates and displacement on a triangle mesh as Phong surfaces. We extend lifted optimization to simultaneously optimize the parameters of the Gaussians while walking on the triangle mesh. SplattingAvatar is a hybrid representation of virtual humans where the mesh represents low-frequency motion and surface deformation, while the Gaussians take over the high-frequency geometry and detailed appearance. Unlike existing deformation methods that rely on an MLP-based linear blend skinning (LBS) field for motion, we control the rotation and translation of the Gaussians directly by mesh, which empowers its compatibility with various animation techniques, e.g., skeletal animation, blend shapes, and mesh editing. Trainable from monocular videos for both full-body and head avatars, SplattingAvatar shows state-of-the-art rendering quality across multiple datasets.