Smartphone-based photogrammetry and videogrammetry versus structured light scanning for the assessment of facial...

This study evaluates the reliability of three-dimensional (3D) facial measurements obtained from different acquisition technologies. Specifically, it investigates whether low-cost smartphone-based photogrammetry and videogrammetry can produce measurements comparable to those derived from structured light scanning, and identifies sources of variability associated with each modality.

Fifteen participants were imaged using three techniques: structured light scanning, smartphone photogrammetry, and smartphone videogrammetry. Inter-landmark distances were computed for each resulting 3D model, including both stable anatomical landmarks and those prone to placement difficulty. A statistical workflow was applied, incorporating non-parametric hypothesis testing, iterative outlier detection, and normality assessments to evaluate measurement variability attributable to acquisition modality, subject movement, and landmark uncertainty.

The three acquisition modalities produced broadly comparable inter-landmark measurements, indicating no substantial loss of accuracy when using smartphone-based methods. However, increased variability was observed in anatomically complex regions, particularly the perioral area, as well as in landmarks known to be difficult to localize. These discrepancies persisted across all modalities, including structured light scanning.

Smartphone-based photogrammetry and videogrammetry represent viable, low-cost alternatives to structured light scanning for 3D facial measurement. Nonetheless, landmark-specific uncertainty remains a critical factor, particularly in challenging anatomical regions. The findings highlight the importance of considering both acquisition modality and landmark reliability in the design and reporting of facial anthropometric studies.