Hochschule Karlsruhe Hochschule Karlsruhe - University of Applied Sciences
Hochschule Karlsruhe Hochschule Karlsruhe - University of Applied Sciences

Intelligent Systems Research Group

End-to-End Workflow for the Generation of Human 3D Scans

Project Work (Bachelor)

Team:
David Milde
Johannes Ott

Supervision:
Prof. Dr.-Ing. Astrid Laubenheimer

Involved on behalf of Q-AMeLia:
Alexander Melde, M.Sc.

Winter Semester 2022/2023

Objective & Results

Within their project work, the students developed and evaluated an end-to-end workflow for recording 3D scans with the scanner of the Laboratory for Machine Vision. The process includes various aspects such as the declaration of consent of the scanned subjects, the calibration of the scanner, the collection of metadata, the scan process itself, and finally the administration and storage of the collected data. The meta data includes informationen such as age, sex, size, weight, bodytype, and detailed clothing information (size, color, texture, accessories) as well as relevant information about the conditions occuring in each scan including detailed clothing information (size, color, texture, accessories), plus a description of the pose, camera angles and lighting conditions.

For the evaluation the process was carried out for a number of fellow students. As a result of the evaluation, a dataset of 66 3D scans of 18 subjects was generated, which will be used for the further evaluation of the 3D scanners performance.

The process will be the basis for larger measurement campaigns.

The scanner is a do-it-yourself scanner and is operated with ImFusion software.