Comparative Study on the Accuracy of Mobile LiDAR Applications in Object-Level Architectural Digitization

The integration of LiDAR (Light Detection and Ranging) sensors in smartphones such as the iPhone 12 Pro Max presents new opportunities for efficient and accurate architectural documentation, particularly at small scales. This study evaluates and compares the scanning accuracy of two mobile LiDAR applications—Polycam Pro and Scaniverse—using a 1:300 scale architectural mockup as the test object. Measurements from each scan were compared against manual reference data to calculate percentage error and assess geometric accuracy. A total of 28 object elements were analyzed across multiple axes. The results show a significant performance gap: Polycam Pro exhibited an average error of 42.58%, while Scaniverse achieved a substantially lower error of 10.36%, with a mean error difference of 32.22%. A paired samples t-test confirmed that this difference is statistically significant (t = 8.09, p = 3.85 × 10⁻⁸), indicating that the disparity is not due to chance. Furthermore, Scaniverse consistently produced more lightweight file sizes, facilitating smoother modeling workflows and digital integration. This study highlights the critical role of software algorithms in mobile LiDAR scanning and suggests that Scaniverse offers greater reliability and accuracy for object-level architectural digitization. The findings contribute to the ongoing development of mobile 3D scanning technologies and support their adoption in precision-driven architectural applications.