Real-time mixed reality head-mounted display occlusion culling method, apparatus, and device

Abstract

Embodiments of the present disclosure disclose a real-time mixed reality head-mounted display de-occlusion method, device and equipment. A specific implementation of the method comprises: generating implicit encoding information through an encoder of a model; performing the following steps through a decoder of the model: determining lower half face deformation information according to third preset feature constraint information, implicit encoding information and lower half face geometry information; generating spatial feature point information set according to mapping correlation weight information, first preset feature constraint information and spatial feature point information set; generating upper half face deformation information according to implicit encoding information, target weight, upper half face key geometry information and upper half face geometry information; changing preset three-dimensional face geometry information to obtain changed three-dimensional face geometry information; and rendering the changed three-dimensional face geometry information to obtain de-occluded face image information. The implementation improves the accuracy and stability of de-occlusion.

Images