Abstract

Issues related to management and workforce play a key role in the productivity gap of construction and manufacturing. Both issues are directly related to the way productivity is measured. Current measurement methods tend to be ineffective because they are labour intensive, costly and prone to human errors whereas they are mainly reactive processes initiated after the detection of a negatively influencing factor. So far, research efforts in automating the measuring process have not achieved full automation because they require prior knowledge of the type of tasks performed in specific working zones. This is highly associated with the lack of depth information. For this purpose, we propose a computationally efficient computer vision method for matching construction workers across different frames based on epipolar geometry and past motion 2D data. The main result of this process is to provide a method for the acquisition of the 4D features (x, y, z, t) that compose the detailed profile of a construction activity in terms of both time and space.