Generating 3D-Consistent Videos from Unposed Internet Photos

Figure 1. Given n unposed input keyframes, the goal is to generate a video of the scene with a realistic camera trajectory and consistent geometry. From top to bottom: Ours, Luma Dream Machine [41] (a commercial video generation model), FILM [50] (a frame interpolation method). Luma hallucinates new buildings (left scene) and statues (right scene) without understanding the scene layout. FILM is unable to handle wide-baseline inputs and produces blurry transitions. See our supplement for video playback.