MIT's Vincent Sitzmann showcases his Scene Representation Group's CVPR research spanning neural scene representations, robotics, and 3D generative models

Projects cover diffusion models, view synthesis, and latent actions