scope:

ABSTRACT

The IEEE 3D Body Processing Industry Connections (3DBP IC) Group develops recommendations that foster interoperability for three-dimensional (3D) body processing, enabling an ecosystem for consumers and creators to use and interchange 3D human body anthropometric information. The 3DBP IC's Quality subgroup focuses on developing preliminary methods, tools, benchmarks, resources, and testing procedures to define and quantify the quality of 3D models as well as the quality of the critical metadata, such as body landmarks and measurements. To further this goal, the 3DBP IC Group completed a comparative analysis of anthropometric methods focused on body measurements and captured 3D body shapes.

The aim of this study is twofold: first, to propose a methodology to assess the compatibility and repeatability of different body measurement extraction methods (e.g., body scanners, smartphone apps, and traditional anthropometry) using consistent criteria, and second, to conduct data collection and the corresponding analyses for the first instance of the application of the proposed methodology.

The experimental study was conducted in two phases. Phase 1 took place in Portland, Oregon (USA) in 2018 and included two booth scanners and four traditional anthropometry measurers. Phase 2 took place in Valencia (Spain) in 2019 and included two booth scanners, four smartphone apps, and two traditional anthropometry measurers. A total of 133 subjects participated in the study. Participants were selected to represent a wide range of body heights and weights. Each subject had 11 body measurements collected twice by each measurer and measuring technology.

The proposed methodology synthesizes: varied, widely used mathematical tools (including statistical models); minimum sample size required for the analyses; and several estimators and plots appropriate for describing the data and quantifying the repeatability and compatibility of the different measurement extraction methods. It provides commonly used criteria to assess most of the statistics. In addition, the methodology includes a procedure to obtain comparable snapshots of the 3D outcomes from digital technologies, facilitating visual assessment by potential data consumers.

This study has gathered a unique dataset that can be used for future research work. The results of the analyses of this dataset constitute useful benchmarks and references for similar studies. All of the participating technologies exhibited a good repeatability for the 11 body measurements considered; however, despite using similar measurement definitions, the compatibility between technologies was low, even for the experts using traditional anthropometry. These results highlight the need for definitions better adapted to 3D imaging systems as well as new methods to assess compatibility. The visual assessment of the 3D objects also shows a positive performance for all digital methods and reveals the differences between the various body scanning solutions and each of the phone apps.

Data and additional resources can be accessed at the IEEE website, specifically measurement data, 3D data, and data from the analyses (i.e., tables, plots, and images).