PandaSet aims to promote and advance the research and development of autonomous driving and machine learning. PandaSet features data collected using forward-looking lidars (PandarGT) and mechanically rotating lidars (Pandar64) with similar image resolution. The collected data is annotated using a combination of cuboid and segmentation annotations (Scale 3D Sensor Fusion Segmentation).

The dataset includes more than 48,000 camera images and 16,000 LiDAR scan point cloud images (over 100 8-second scenes). The dataset also includes 28 annotations for each scene and 37 semantic segmentation labels for most scenes.