CPC G06T 7/70 (2017.01) [G06N 3/045 (2023.01); G06V 10/46 (2022.01); G06V 10/757 (2022.01); G06V 10/762 (2022.01); G06V 10/82 (2022.01); G06T 2207/10028 (2013.01)] | 16 Claims |
1. An object positioning method, comprising:
receiving 3D point cloud data obtained from a scene comprising a target object by a sensing device;
extracting a key point from the 3D point cloud data by a processing device;
inputting surrounding area data centered on the key point and a preset feature descriptor of the target object to a neural network by the processing device to calculate a scene feature descriptor of the scene, wherein the preset feature descriptor is obtained from a 3D standard template of the target object, and the neural network is trained by a database including scenes composed of 3D point clouds;
performing feature matching between the scene feature descriptor and the preset feature descriptor by the processing device; and
calculating a position of the target object in an actual space by the processing device.
|