Augmented reality occlusion method based on image segmentation and custom layer method

Abstract

The invention discloses an augmented reality shielding method based on image segmentation and the customized layer method. The method mainly comprises steps of acquiring prior knowledge, self-defining a space position relation of each layer in a scene according to the acquired prior knowledge and carrying out corresponding content filling; extracting a true shielding object; and carrying out shielding rendering. According to the invention, the defect of quite poor timeliness of a shielding method of depth calculation in an operation process in the prior art can be overcome; the defect that three-dimensional modeling and coverage completion of true objects are mainly achieved depending on manual setting of markers in a model reconstruction-based shielding method due to the high difficulty are overcome; and it is achieved that difficulty in effectively separating a background in a real scene from a to-be-shielded foreground target is reduced and the method is characterized by the separation accuracy.

Description

technical field [0001] The invention relates to the technical field, in particular to an augmented reality occlusion method based on image segmentation and a custom layer method. Background technique [0002] In the augmented reality scene, the occlusion relationship between virtual and real objects can be divided into two categories: (1) the occlusion of the virtual object to the real scene; (2) the occlusion of the real object to the virtual scene. The former effect is achieved in most augmented reality systems. In addition to the existence of real objects in the real scene, the augmented reality scene with the real background as the background must also be considered, so when realizing the occlusion of the virtual scene by the real object, how to effectively combine the background in the real scene with the occluded foreground target Separation has become a research focus. Among the virtual and real occlusion methods proposed so far, they can be roughly divided into thr...

AI Technical Summary

Problems solved by technology

[0007] In the process of realizing the present invention, the inventor found that there are at least defects in the prior art, such as the problem that it is difficult for the traditional mobile terminal to obtain scene depth information.

Images