The accuracy of a three-dimensional face model reconstructing method based on conventional clinical two-dimensional photos

Abstract

Background This study aims to investigate the accuracy of a three-dimensional (3D) face reconstruction method based on conventional clinical two-dimensional (2D) photos. Methods Twenty-three patients were included, and Character Creator v3.2 software with the Headshot v1.0 plugin was used for 3D face model reconstruction. Various facial landmarks were finely adjusted manually to refine the models. After preprocessing and repositioning, 3D deviation analysis was performed. The accuracy of the landmarks in different dimensions was determined, and twelve facial soft tissue measurements were compared to validate the clinical potential of the method. Result The reconstructed 3D face models showed good facial morphology with fine texture. The average root mean square errors between face scan models and reconstructed models at perioral area (1.26 ± 0.24 mm, 95%CI: 1.15–1.37 mm) were significantly smaller than the entire facial area (1.77 ± 0.23 mm, 95%CI:1.67–1.88 mm), P < 0.01. The deviation of menton of soft tissue was significantly larger than pronasale (P < 0.01). The deviations of all landmarks in the Y-direction were significantly larger than those in the other 2 dimensions (Y > Z > X, P < 0.01). A significant difference (P < 0.05) of approximately 1.5 mm was found for facial height. Significant differences (P < 0.05) were also identified in the remaining 6 soft tissue measurements, with average deviations no greater than 0.5 mm (linear measurement) or 1.2° (angular measurements). Conclusion A 3D face modeling method based on 2D face photos was revealed and validated. The reconstruction accuracy of this method is clinically acceptable for orthodontic measurement purposes, but narrow clinical indications and labor-intensive operations remain problems.

Digital Human: Three-dimentional face model reconstruction based on Hifi3dface model with clinical two-dimentional photos

“…In addition, due to the plumpness of soft tissue in the zygomatic region and the lack of clear soft tissue landmarks, depth information in this area was di cult to match accurately when using a lateral photo, which was another reason for the poor reconstruction in the cheek. Mao [14]et al used 3D modeling software for manual face remodeling, which yielded higher precision than this method. However, their approach was complex, and the reconstruction outcome was dependent on the experience of practitioners.…”

“…Acquisition of 2D photographs and 3D scans 2D photo and 3D data acquisition methods were consistent with Mao [14], and they were obtained on the same day. FaceSCAN3D System (3DShape, Erlangen, Germany) was used to capture the 3D model.…”

“…After repositioning, 3D model registration and deviation analysis were performed by Geomagic Control (Geomagic, Morrisville, North Carolina, USA). The alignment registration method was consistent with Mao [14]. First, the bilateral canthus, Pronasale and soft tissue nasion were marked manually for pre-alignment.…”

AI and Face-Driven Orthodontics: A Scoping Review of Digital Advances in Diagnosis and Treatment Planning

“…The process of creating 3D faces from 2D pictures is validated, as the acquired measurements are clinically acceptable. Nonetheless, this process is time-and labour-demanding [100].…”

Optimization of a Novel Automated, Low Cost, Three-Dimensional Photogrammetry System (PHACE)

“…Prior studies have only evaluated the dimensional accuracy by measuring displacement of rendered models to a reference. 38,39 However, measuring differences in linear distance between rendered and reference models is limited to interpreting 3D volumetric changes from a 2D imaging plane. Therefore, we quantitatively measured volumes to objectively assess dimensional and morphological accuracy.…”