To comprehensively and accurately evaluate the severity degree of asphalt pavement shoving, as well as realize the automatic calculation of the corresponding three-dimensional indicators, the indoor three-dimensional laser detection equipment was used to obtain high-precision and high-density surface laser point cloud data of shoving models, and the Lowess algorithm was adopted for data preprocessing. The RANSAC algorithm was then utilized to extract the contour data of the bottom of the shoving0, and the three-dimensional model of the shoving was established based on the HARR matrix. On this basis, the height-difference method and the micro-element approximation algorithm were used to calculate the three-dimensional indicators such as the maximum height, bottom area, uplift volume, and the maximum slope of the driving direction of the shoving. The calculation results of the three-dimensional indicators of two severity levels of light and heavy were compared, and the influence of laser line longitudinal spacing on calculation error was studied. Results show that when the longitudinal spacing of the laser line was 0.5 mm, the calculated relative errors of the maximum height, bottom area, uplift volume, and the maximum slope of the driving direction were less than 3.28%, 2.17%, 3.76%, and 1.97%, respectively. As the spacing increased to 10 mm, the relative error of the three-dimensional indicators of the shoving gradually increased due to the partial missing of the shoving model. When the spacing was 5 mm, the maximum relative error of the three-dimensional indicators was 3.49%, 4.65%, 7.11%, and 7.27%, respectively. When the spacing was 10 mm, the relative errors of the uplift volume and the maximum slope of the driving direction both exceeded 10%. It is recommended that when using three-dimensional laser technology to measure shoving, the longitudinal spacing should be not more than 5 mm to ensure the accuracy of the calculation results of the three-dimensional indicators. The three-dimensional indicators calculation method can evaluate the severity of the shoving and assess traffic safety risk.