Review of indoor real scene 3D reconstruction technology
A semantically rich,geometrically accurate,and topologically complete indoor 3D model poses a significant challenge in the construction of a realistic 3D representation of China.It holds crucial application value in various fields such as indoor navigation and location services,virtual reality,and smart homes.However,the intricate layout of indoor spaces,diverse entity elements,and occlusions present substantial challenges to the 3D reconstruction of indoor scenes.In recent years,the reconstruction of 3D indoor real scenes has garnered widespread attention,yet a systematic summary of state-of-the-art techniques is lacking.Therefore,this paper aims to provide a comprehensive summary of state-of-the-art techniques related to 3D indoor real scenes.Firstly,mainstream methods for collecting 3D indoor space data are briefly outlined.Subsequently,delving into the key techniques involved in the reconstruction process,this paper reviews state-of-the-art methods,highlighting their advantages and disadvantages in terms of semantic recognition and classification of entity elements,geometric model generation of entity elements,as well as the organization and reprentation of spatial topological models.Finally,the paper concludes with a summary of challenges associated with the reconstruction of 3D indoor real scenes,followed by a discussion on future research trends.
3D indoor real scenedata registrationclassification of entity elementsspatial topology modelgeometric model of entity elements
NETL
NSTL
万方数据
