AR scene application method based on single picture positioning

Abstract

The invention discloses an AR scene application method based on single picture positioning. The method comprises the steps that a server performs three-dimensional map modeling on image data of a global scene; wherein the three-dimensional map modeling comprises a descriptor corresponding to each feature point of the image data and a three-dimensional coordinate parameter corresponding to the descriptor; the mobile terminal acquires single picture data and uploads the single picture data to the server; the server performs feature point extraction on the single picture data to obtain a descriptor corresponding to each feature point; the descriptor corresponding to each feature point in the single picture is matched with the descriptor corresponding to each feature point of the global scenemodel, and the pose of the single picture in the global scene model is obtained through the three-dimensional coordinate parameters corresponding to the descriptors; and the pose of the single picturein the global scene model is converted into an AR application coordinate system, and the AR application is activated.

Description

technical field [0001] The present invention relates to the technical field of AR image recognition, in particular to an AR scene application method based on single picture positioning. Background technique [0002] Image-based modeling technology has always been an important research direction in the field of computer vision. At present, the main process is to establish the connection between images through feature point extraction and feature point descriptor matching, and then use 3D vision and mathematical modeling techniques for the physical world to build a real-world model from the image. The disadvantages are: the camera parameters and models of pictures from various sources are different, and the estimation error of the camera model leads to systematic errors; and the amount of image data is too large, resulting in too long calculation time (ranging from a few days to a week). [0003] At present, AR (augmented reality) technology is developing rapidly. Mobile phon...

AI Technical Summary

Problems solved by technology

The disadvantages are: the camera parameters and models of pictures from various sources are different, and the estimation error of the camera model leads to system errors; and the image data is too large, resulting in too long calculation time (ranging from a few days to a week)

The disadvantage is that the current AR applications are concentrated in a small area, and are limited by plane recognition and image recognition, so it is impossible to realize AR in the pre-established large scene model

Images