1 paper
Chensheng Peng, Ido Sobol, Masayoshi Tomizuka +3
We present a novel framework for training 3D image-conditioned diffusion models using only 2D supervision. Recovering 3D structure from 2D images is inherently ill-posed due to the…