World-consistent Video Diffusion with Explicit 3D Modeling

Qihang Zhang, Shuangfei Zhai, Miguel Ángel Bautista Martin, Kevin Miao, Alexander Toshev, Joshua M. Susskind, Jiatao Gu

摘要

Figure 1. WVD predicts 6D videos from an image, unifying various 3D tasks with a single diffusion model.