"There is a lack of viable methods to map and label existing infrastructure. This is one of the grand challenges of engineering in the 21st century noted by the “Restoring and Improving Urban Infrastructure” report of the US National Academy of Engineering (NAE). For instance, over two thirds of the effort needed to model even simple infrastructure is spent on manually converting a cloud of points to a 3D model. The result is that only very few constructed facilities today have a complete record of as-built information and that as-built models are not produced for the vast majority of new construction and retrofit projects, which leads to rework and design changes that cost up to 10% of the installed costs.This project plans to test whether a novel framework proposed by the researcher can reasonably detect and classify common building objects from visual and spatial data, for the purpose of significantly reducing the time it takes to create the as-built geometric Building Information Model (BIM) of an existing facility. Under the proposed plan of work, the visual characteristics of civil infrastructure element types are identified and numerically represented using image analysis tools. The derived representations along with their inferred relative topology are then used to form the element parts for learning the element category models. These models are used to automate the detection of element types and their local poses from arbitrary views. The detected elements, by further estimating their distance to the observer and 3D bounding box, are mapped onto the 3D point clouds rendered with colour and texture. If successful, this project will provide the research community with the first view and scale-invariant, civil infrastructure object detection method that is capable of automatically quantifying object parts for training, detecting objects from arbitrary viewing points, and estimating the layout of the objects in the 3D physical space."