You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi,I wonder when you calculate FVD metric, do you generate videos conditioned on an initial RGB image from dataset, or you just use 3d box and hdmap infomation as condition and generate videos from pure noise?