Preprint
Aug 2026
DreamTraj: Generating 6-DoF Object Trajectories by Reading Unrendered Video Diffusion Latents
Accurate prediction of object trajectories during manipulation is essential for closing the perception-action loop. Progress is limited on two fronts: available datasets lack fine-grained language-to-motion annotations, and existing predictors either rely on privileged inputs such as video, depth, or CAD models, or rec...
T. Ding, Zhen Luo, Yi-Xuan Yang et al.
· 0 citations