Near Real-Time 3D Reconstruction of Construction Sites Based on Surveillance Cameras

The 3D reconstruction of construction sites is of great importance for construction pro-gress, quality, and safety management. Currently, most of the existing 3D reconstruction methods are unable to conduct continuous and uninterrupted perception, and difficult to achieve registration with real coordinates and dimensions. This study proposes a hierar-chical registration framework for 3D reconstruction of construction sites based on surveil-lance cameras. This method can quickly perform on-site 3D reconstruction and restoration by taking fixed views such as surveillance camera images as inputs without transfer learning and camera calibration. This framework can complete the 3D point cloud esti-mation and registration of construction sites within an average of 3.105s through surveil-lance images. The average RMSE of the point cloud within the site is 0.358m, which is better than most point cloud registration methods. Through this method, 3D data within the scope of surveillance cameras can be quickly obtained, and the connection between 2D and 3D can be effectively established. Combined with visual information, it is beneficial to the digital twin management of construction sites.