This is a compact, viewer-friendly sample extracted from a much larger HQ locomotion export generated by the QualityVision Motion Dataset Engine. Looking for the full commercial export or custom delivery? See pricing & ready-made bundles on qvision.space/dataset-pricing.

What’s inside

data.jsonl: one JSON object per line (one frame per row) with 33 MediaPipe/BlazePose landmarks (x,y,z… See the full description on the dataset page: https://huggingface.co/datasets/Alaaharoun/QualityVision-Locomotion-Pose-Dataset-Walking-Jogging-Running.