The road toward smart city infrastructures: A review on 3d facade reconstruction using images

Abstract

Numerous types of solid structures protect coastal activities and cities from the damaging effects of tides and waves all around the globe. Thus, having a three-dimensional digital representation of the physical environment would help decision-makers in understanding the dynamic nature of coastal environments and implementing effective mitigation strategies. Generally speaking, Feature matching, Structure from Motion (SFM) and Multi-View Stereo (MVS) algorithms are used in this order to achieve realistic results. The Literature shows that there is a constant evolution of new techniques and technologies either with learning based or hand-crafted approach, which gives a possibility to integrate different method to optimize each step of the three-dimensional reconstruction process. The aim of this paper is to present the progress of three-dimensional modelling methods that use ground-level images by providing an overview of the latest applications and a comparison of their results. Overall, the state-of-the-art in three-dimensional building modelling using ground-level imagery is rapidly evolving, and new ways are being developed to improve the efficiency, accuracy and scalability of the process