Estimating somatotype from a single-camera 3D body scanning system

To what extent do somatotype structures affect athletic performance in professional athletes?

BACKGROUND

This study aimed to ascertain the influence of somatotype structures on athletic performance among professional athletes.

METHODS

Twenty-eight professional athletes, comprising 14 handball players and 14 volleyball players, were enlisted as participants for this study. As per the methodology, evaluations encompassing explosive power, strength, flexibility, horizontal jump, agility, sprint, and hand-foot reaction tests were conducted. Following physiological assessments, parameters including body mass index, body fat percentage, and somatotype were computed. The impact of athletes' somatotype structures on their athletic performance was investigated employing linear regression analysis.

RESULTS

The endomorphy value serves as a negative predictor (r2=0.159) for explosive power, whereas the mesomorphy (r2=0.204) and ectomorphy (r2=0.379) values act as positive predictors of explosive power. Ectomorphy value negatively predicts grip strength (r2=0.399) and back strength (r2=0.214), while mesomorphy value positively predicts grip strength (r2=0.233) and back strength (r=0.338; r2=0.114). Mesomorphy value negatively predicts flexibility (r2=0.247), long jump (r2=0.311), sprint (r2=0.256), and reaction (r2=0.133), whereas ectomorphy value positively predicts flexibility (r2=0.225), long jump (r2=0.463), sprint (r2=0.393), and reaction (r2=0.208).

CONCLUSIONS

This study showed a correlation between various somatotype structures and indicators of athletic performance. Within this framework, it can be inferred that mesomorphy and ectomorphy values serve as reliable predictors of athletes' performance. Furthermore, individuals with elevated mesomorphy values may exhibit enhanced strength performance, while those with higher ectomorphy values may demonstrate superior flexibility performance. Hence, when designing training programs aimed at enhancing athletes' performance attributes, it is imperative to account for the influence of body type on such performances.

Modification and refinement of three‐dimensional reconstruction to estimate body volume from a simulated single‐camera image

Objective

Body volumes (BV) are used for calculating body composition to perform obesity assessments. Conventional BV estimation techniques, such as underwater weighing, can be difficult to apply. Advanced machine learning techniques enable multiple obesity-related body measurements to be obtained using a single-camera image; however, the accuracy of BV calculated using these techniques is unknown. This study aims to adapt and evaluate a machine learning technique, synthetic training for real accurate pose and shape (STRAPS), to estimate BV.

Methods

The machine learning technique, STRAPS, was applied to generate three-dimensional (3D) models from simulated two-dimensional (2D) images; these 3D models were then scaled with body stature and BV were estimated using regression models corrected for body mass. A commercial 3D scan dataset with a wide range of participants (n = 4318) was used to compare reference and estimated BV data.

Results

The developed methods estimated BV with small relative standard errors of estimation (<7%) although performance varied when applied to different groups. The BV estimated for people with body mass index (BMI) < 30 kg/m² (1.9% for males and 1.8% for females) were more accurate than for people with BMI ≥ 30 kg/m² (6.9% for males and 2.4% for females).

Conclusions

The developed method can be used for females and males with BMI < 30 kg/m² in BV estimation and could be used for obesity assessments at home or clinic settings.