Starting with an innovative approach for generating 4D sewing patterns, OpenDress was faced with the task of developing a pipeline that leads from an autonomous user scanning process to the development of a personalized avatar to a production procedure for 4D sewing patterns. We were looking for an approach that makes it possible to reconstruct a 3D scan from a 2D photo (see also : Ballester et. al.(2018), Saito et. al. (2019), Saito et. al. (2020), Zhang et. al. (2020) apply AI trained movement to the generated avatar and construct the sewing pattern directly on the movement-trained avatar. The goal was to develop a pipeline, i.e. as platform-independent as possible, without special hardware for broad and easy user access. This paper presents partial results of a larger validation study that compared data results from different low-threshold and pseudo-monocular technologies to investigate the development of an ai-driven approach. The study conducted body measurements of more than 200 participants, using the Lidar sensor scan function from mobile devices and photos generated via mobile cameras in comparison to scans from the structure scanner via tablet and measurements by hand. Shape and pose parameters from the systems tested were analyzed in order to create a customized, parametric 3D model (avatar). Here the importance of physical realistic shape-, pose, and measurement parameters of the individual human body was emphasized, in order to further process the 4D sewing patterns.