Building Schematic Analyzer

Construction companies often overestimate or underestimate the amount of materials needed for construction. This costs the company money in extra delivery fees and unused material. Our project aims to solve this problem by accurately outputting the quantity of materials needed and the associated cost through an image of the building’s framing schematic. In order to achieve this goal, we are using a convolutional neural network to train a model to recognize key features such as walls, windows, doors, and posts.  Once these features are extracted, we will use another algorithm to determine the quantity of materials needed and the cost.