ABSTRACT
The exponential increase in the volume of images and videos captured in the built environment with smartphones, fixed cameras, unmanned aerial and ground vehicles, and laser scanning devices, provides a unique opportunity to digitally record and analyze the entire construction life cycle. To make these ambient big visual data actionable, research has developed and applied state-of-the-art computer vision techniques to produce 3D reality models of ongoing operations and automatically organize and manage them over project timelines. The integration of these models with Building Information Modeling (BIM) and project schedules has enabled development and use of many new solutions for progress monitoring, safety inspection, quality assessment/control, productivity analysis, operation maintenance, and building energy performance. In this chapter, we present how the cutting-edge computer vision and machine learning techniques such as deep learning, object detection and Structure from Motion together with BIM and schedule are integrated and applied in the context of built environment to enable performance monitoring. For each use case we will provide a concise literature review and assessment of current state-of-the-art solutions in the market, and will discuss the key underlying methods and recent solutions in detail. We will demonstrate their real-world performance by using multiple building and infrastructure project case studies. We will also discuss the tangible benefits in forms of Return on Investment and will share lessons learned across these projects. The challenges of applying these techniques in typical project workflows and open areas for research and development are also discussed in detail.
