An augmented reality virtual-real occlusion processing method based on an active appearance model

Abstract

The invention requests to protect an augmented reality virtual and real occlusion processing method based on an active appearance model. A non-rigid body tracking technology based on an active appearance model and a virtual-real occlusion processing technology based on depth calculation are used. The method specifically comprises the following steps: 1. The method comprises the steps of 1, aligning shape samples in a training sample set, 2, constructing an average shape model: constructing an average shape model; 3, construction of a texture model; 4, construction of an active appearance model: after construction of the shape model and the texture model is completed, any new non-rigid object input image can be represented by shape model parameters bs and texture model parameters bg, and the shape model and the texture model are combined to establish the active appearance model; And 5, model matching and non-rigid object segmentation. According to the method, a good shielding processingeffect can be achieved, and occurrence of 'shadows' is avoided.

Description

technical field [0001] The invention belongs to the technical field of augmented reality, and in particular relates to an augmented reality virtual and real occlusion processing method based on an active appearance model. Background technique [0002] Most of the current research discusses the occlusion between rigid objects and virtual objects, which is inconsistent with the actual situation. For example, hands entering the observation field, other operators, etc. are all non-rigid objects. Unlike rigid objects, non-rigid objects are prone to irregular deformation, internal folds and distortions during motion. Therefore, how to deal with the virtual-real occlusion relationship for non-rigid objects needs further research. [0003] Fuhrmann et al. studied the occlusion processing of three types of real objects and virtual objects: static objects, moving rigid objects and moving non-rigid objects. For non-rigid objects, the human arm is regarded as a multi-joint rigid objec...

AI Technical Summary

Problems solved by technology

The method based on depth calculation is currently the main method for dealing with the virtual and real occlusion of moving non-rigid objects, but most of these methods have a large amount of calculation and real-time performance is difficult to guarantee

How to track and segment non-rigid objects in the image, by only comparing the depth value between the segmented area and the virtual assembly guidance information, to avoid the problem of large amount of calculation caused by the depth value judgment on the entire image, further research is needed Research

Images