Open MarvinKlemp opened 1 year ago
Are you using the perception dataset? Our bounding boxes should have 3D info including heading here: https://github.com/waymo-research/waymo-open-dataset/blob/c9bf4d829bdf10f48a0dec2ef0ef2e5e05a8a0db/src/waymo_open_dataset/label.proto#L38
Is there a way to get the heading of the objects using the 2D bounding boxes (vehicles)?