Human-computer interaction method and device based on panoramic view

Abstract

The invention discloses a human-computer interaction method and device based on a panoramic view. The method includes: receiving an operation on a panoramic view image triggered by a user on a vehicle screen, and obtaining corresponding operation information; the operation information includes at least one screen coordinate; according to The operation information displays at least one driving function for the user to select; converts at least one screen coordinate included in the operation information to obtain at least one corresponding world coordinate; controls the vehicle to perform the corresponding driving according to at least one world coordinate and the driving function selected by the user Function. The present invention solves the problem that the existing general panoramic view system cannot directly interact with the user at the driving level. By positioning and converting the screen coordinates selected by the user, the screen coordinates displayed on the screen plane and the 3D space in the real world are realized. Corresponding to the world coordinates, it is convenient for the user to control the vehicle based on the displayed panoramic view image to complete the corresponding driving function.

Description

technical field [0001] The invention relates to the field of artificial intelligence, in particular to a method and device for human-computer interaction based on panoramic viewing. Background technique [0002] The panoramic surround view system integrates information from on-board sensors such as cameras and radars (such as millimeter-wave radars, ultrasonic radars, etc.), and covers up to 360-degree real scenes around, including auxiliary information such as lane lines and object frames. A system that is displayed on the screen of the vehicle for the user to view the surrounding conditions of the vehicle. The virtual visualization system is a system that displays lane lines, objects, signage and other information in the form of a virtual model on the screen of the vehicle for users to view by fusing the information of the vehicle sensors. At present, it is mainly used to display the situation ahead of the road . [0003] The 2D or 3D surround view system of the real sce...

AI Technical Summary

Problems solved by technology

[0003] The 2D or 3D surround view system of the real scene can only display the environment around the car for the user to view, but cannot directly interact with the user in the real world based on the visual images in the surround view system

The virtual visualization system can perform some interactions, such as parking space selection in the process of automatic parking, but it cannot display real image information, and the field of view is not wide enough, which increases the difficulty for users to communicate with it and limits its scope of application

[0004] In addition, in the transition stage from assisted driving to automatic driving, many functions can only solve part of the problem, and the driver's intention still plays an important role in driving.

Simple scenarios such as changing lanes, overtaking, and turning can be performed simply by buttons or simple commands, but it is difficult to clearly and accurately describe the user's driving intentions, so that for complex intentions, instruction stacking is required to realize, or it is difficult to realize

For example, following a vehicle, the current adaptive cruise function can only follow the vehicle directly ahead. If you want to follow other vehicles, the existing technology has significant limitations in implementation and interaction; In the specified position, the existing technology also has great limitations in the interactive realization

Images