Multiple-choice questions about how objects move, derived exactly from dense
6-DoF (object→camera) pose trajectories rather than guessed from pixels. Each row pairs
a short 6fps video clip with one motion MCQ, its per-second motion trajectory, and the
whole-video aggregated answer. The intended task: watch the clip and pick the motion that
actually happens.
Built with the motion-qa pipeline… See the full description on the dataset page:
https://huggingface.co/datasets/livctr/bop-motion-mcq.
