Despite many safety considerations addressed in lift pre-planning, the ability to provide real-time safety assistance to crane operators and to mitigate human errors during the lifting operation is missing. This research developed a framework for an operation-level safety assistance system enabled by continuous crane motion capturing, pro-active hazard analysis, and real-time 3D visualization. A sensor system captures the critical motions of crane parts, based on which the entire crane motions are reconstructed in real-time with as-is lift site conditions modeled based on point cloud data. The risk of colliding the crane parts and lifted load to surrounding obstructions is analyzed and warnings are provided to the operator through a user interface. A prototype system was developed based on the framework and deployed on a mobile crane. Field test results indicate that the system can accurately reconstruct crane motion in real-time and provide pro-active warnings that allow the operator to make timely decisions to mitigate the risk.