[ Back ]
Instant Gaussian Splatting Generation for High-Quality and Real-Time Facial Asset Rendering
Dafei Qin, Hongyang Lin, Qixuan Zhang, Kaichun Qiao, Longwen Zhang, Jun Saito, Zijun Zhao, Jingyi Yu, Lan Xu, Taku Komura
IEEE Transactions on Pattern Analysis and Machine Intelligence ยท 2025
Abstract
We propose GauFace, a Gaussian Splatting representation tailored for efficient facial asset rendering, together with TransGS, a diffusion transformer that generates GauFace assets from meshes, textures, and lighting conditions. The method targets both high visual quality and real-time performance across desktop and mobile platforms.
Imported from Google Scholar and refined for the site profile page.