Abstract
In recent years, advances in computer hardware, graphics rendering algorithms and computer vision have enabled the utilization of 3D building reconstructions in the fields of archeological structure restoration and urban planning. This paper deals with the reconstruction of realistic 3D models of buildings façades, in the urban environment for cultural heritage. The proposed approach is an extension of our previous work in this research topic, which introduced a methodology for accurate 3D realistic façade reconstruction by defining and exploiting a relation between stereoscopic image and tacheometry data. In this work, we re-purpose well known deep neural network architectures in the fields of image segmentation and single image depth prediction, for the tasks of façade structural element detection, depth point-cloud generation and protrusion estimation, with the goal of alleviating drawbacks in our previous design, resulting in a more light-weight, robust, flexible and cost-effective design.
Highlights
The constant differentiation of the number and architectural characteristics of buildings in a city, has highlighted the necessity of building documentation in order to better organize, plan and control the structural specifications of each city
An ideal automated 3D building façade reconstruction approach designed for the purpose of providing photo-realistic 3D building façade reconstructions for building documentation, should exhibit robustness and flexibility against different design characteristics and be as much as resource and computational cost-efficient as possible
The contributions of the present paper lie in eliminating the limitations of our previous approach. (a) By introducing an auto-encoder neural network architecture for depth estimation using a single RGB image instead of a stereoscopic image sensor rig design, which can potentially increase the flexibility and decrease the overall cost of our framework. (b) By incorporating a deep learning-based façade segmentation stage based on generative adversarial networks, enabling for more scalable and robust façade element detection. (c) By integrating computational geometry techniques and point cloud processing algorithms to produce a detailed reconstructed 3D surface, enhancing the automation and adaptability of the suggested workflow
Summary
The constant differentiation of the number and architectural characteristics of buildings in a city, has highlighted the necessity of building documentation in order to better organize, plan and control the structural specifications of each city. The generation of 3D photo-realistic 3D building façade reconstructions is a challenging task since building façade designs vary in the number and architectural complexity of their structural components. This attribute becomes a severe issue in older cities in which historical and modern building designs co-exist, encapsulating decades of architectural design trends. An ideal automated 3D building façade reconstruction approach designed for the purpose of providing photo-realistic 3D building façade reconstructions for building documentation, should exhibit robustness and flexibility against different design characteristics and be as much as resource and computational cost-efficient as possible
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
