Method and device for determining three-dimensional pose of object and electronic equipment

Abstract

The embodiment of the invention provides a method and device for determining a three-dimensional pose of an object and electronic equipment. The method comprises the following steps: acquiring a target image which is acquired by a first camera and comprises a target object, and inputting the target image into a pre-trained deep neural network model for detection, obtaining a target coordinate point in the target image, and determining the three-dimensional pose of the handle relative to the first camera coordinate system according to the target coordinate point and the parameters of the handlein the world coordinate system. As the imaging coordinate points of the handle of the target object at the two ends of the target image can be obtained through the deep neural network model, according to the coordinate point and the parameters of the handle in the world coordinate system, the three-dimensional pose of the handle with any pose relative to the first camera coordinate system can beaccurately determined, while an object with the handle does not need to be fixed through a fixing clamping groove, so that the flexibility of subsequent target object grabbing is greatly improved.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a method, device and electronic equipment for determining a three-dimensional pose of an object. Background technique [0002] In the field of artificial intelligence technology, the technology of grasping objects through robotic arms is widely used. For example, in scenarios such as chat robots, service robots, and robotic arms loading goods, the technology of grasping objects with robotic arms is required. [0003] In the process of grabbing objects by the robotic arm, it is first necessary to determine the three-dimensional pose of the object, that is, the three-dimensional position and three-dimensional pose of the object, and then determine the grasping scheme according to the three-dimensional pose of the object and carry out the grasping of the object. In some application scenarios, the robotic arm needs to grasp objects with handles, such as coffee cups, m...

AI Technical Summary

Problems solved by technology

It can be seen that in the above method, the determination of the three-dimensional pose of the object does not achieve intelligence, resulting in that the robotic arm can only grasp the object with a handle fixed by the fixed card slot, and the flexibility of grasping is very low.

Images