Abstract: Based on the point cloud data and aerial photos of a university library, the 3D model is constructed by using 3ds Max and Smart 3D. On the one hand, the method of human-computer interaction is adopted for modeling, and the 3D laser point cloud data is imported into 3ds Max software to establish the 3D digital model. On the other hand, the semi-automatic manual mode is used for modeling, and the Smart 3D software is used to model the building entity after the aerial camera data is encrypted, automatically modeled, mapped and other operations. The characteristics and differences of the two modeling methods are compared in terms of modeling efficiency, model aesthetics, modeling accuracy and applicable objects. The research shows that 3ds Max is more suitable for regular body and high-precision modeling, while the efficiency of modeling is relatively low; Smart 3D modeling is more efficient and more suitable for rapid modeling in a large range of scenes. The conclusions are helpful to the selection of modeling methods under different requirements, and provide a reference for the selection of 3D modeling methods for research, engineering and experiment.

Key words: city 3D model, modeling efficiency, modeling accuracy, 3ds Max, Smart 3D