Translations:Diffusion Models Are Real-Time Game Engines/81/en

Neural methods for reconstructing 3D representations have made significant advances over the last years. NeRFs (Mildenhall et al., 2020) parameterize radiance fields using a deep neural network that is specifically optimized for a given scene from a set of images taken from various camera poses. Once trained, novel points of view of the scene can be sampled using volume rendering methods. Gaussian Splatting (Kerbl et al., 2023) approaches build on NeRFs but represent scenes using 3D Gaussians and adapted rasterization methods, unlocking faster training and rendering times. While demonstrating impressive reconstruction results and real-time interactivity, these methods are often limited to static scenes.