Design and Implementation of Virtual Real Fusion Metaverse Scene Based on Deep Learning

Wenying Zeng, Jianhua Hu, Lixia Huang, Weimei Wu

Abstract: With the rapid development and application of metaverse hotspot technology, it is necessary to simulate and construct various virtual scenes, which may be difficult to meet complex application requirements. To solve the problem of scene construction as above, relevant technical research has been conducted, and a fast method for constructing virtual scenes has been designed. Based on deep learning algorithms, a 3D digital space is constructed through virtual 2D characters and landscape maps to reproduce various real scenes; At the same time, it can interact, roam, update scenes, engage in fitness, socializing, and more. Realized the intelligent scene construction and fusion of virtual and real fusion in the metaverse. This provides a design and implementation reference for the generation of scenes in the metaverse.