Construction progress is a key indicator throughout the entire project construction process and is of great significance to project management and control. Traditional construction progress tracking mainly relies on manual records, which is time-consuming, labor-intensive, error-prone, and difficult to meet the needs of efficient project management and control. This study combines Building Information Modeling (BIM) technology with three-dimensional reconstruction technology to study and verify an automatic construction progress tracking method based on digital twins. Based on the current data collection of 3D reconstruction equipment, the study first introduces a 3D point cloud semantic segmentation model to automatically identify the main engineering components, thereby realizing rapid registration of the 3D point cloud and BIM model to build a digital twin model of the construction site; at the same time, it is proposed based on The construction progress judgment method of BIM and point cloud overlap rate enables rapid extraction of construction progress. Finally, the feasibility of this research method is verified by taking the fourth-floor scene of a newly built building as a case.