Design of a 3D Headscanner

Design of a 3D Headscanner: Facilitating the design of ultra-personalized products for the differently-abled.

Gupta, Yaman (TU Delft Industrial Design Engineering)

Song, Y. (mentor)
Ahsmann, L.L. (graduation committee)

Delft University of Technology

Integrated Product Design

2021-08-26

The 3D Head-scanner is designed with the intent to scan a customer’s head such that the client can design personalized glasses using the generated mesh. The uniqueness of this process is due to the fact that the customers are uniquely disabled and therefore unable to clearly announce their comfort levels. The goal of this project is to design a new Head-scanner for Maat! since the process that they currently use involves interacting a lot of times with customers who move around a lot making scans invalid and operators having to put in extra effort. The client therefore wants a new and improved approach to scanning, one that is faster, accurate, portable and comfortable for their customers. The initial brief was to replace their existing workflow through this new design, however, after some research and discussion, the approach was changed to suit Maat’s future strategy of setting up stationary scanners all around the Netherlands in locations such as Down polis or Community centers where people could come for checkups and have themselves scanned as well resulting in a significant cut-down on the clients’ travelling time. The project starts with some contextual study where the clients are interviewed on their process, their observations and their expectations from the product. Further, clients are shadowed on a number of their customer visits for observation and gaining a first-hand understanding of a typical scanning process, customer behavior and interaction, involvement of parents, environment and noting down certain areas of interest that could motivate insights. The observations lead to a deeper understanding of a child-customer’s behaviour including points of distraction,
various approach strategies employed by operators and how these aspects
could be leveraged outside of a product’s workflow. Certain important points to note are that children have to be distracted at a common point for some amount of time since the scanning process takes some time. This could either be a parent
standing in front of them or their favourite show on a phone. Due to the motion of the scanner, kids often get distracted towards the operator leading to parents often having to hold their head straight. Technological research is the next step
and this involves looking at market competitors, their price ranges, techniques
employed and feasibility with regards to the current context. Similarly, a number of scanning techniques are also considered before photogrammetry is eventually
selected due to its speed, accuracy, ease of availability and pricing. Further tests are also carried out that involve comparing photogrammetry with structured light, scanning dummy heads for accuracy and working with the coding aspect and relevant software. These steps directly inform certain design decisions which serve as constraints based on which concept ideas for the embodiment and possible look and feel are ideated upon. Factors such as area of capture, landmarks and available space influence the design of the product as it
undergoes a number of iterations before settling on the current version. The current Head-scanner makes use of 3 cameras to capture the subject’s face
along with the requisite landmarks in less than a second, with only the click of
a button. As it is connected to a laptop, all the post-processing happens on the
system where the different camera views are aligned creating a complete head. This head is then showed to the operator on a Viewer for them to check.
The final Prototype is repeatedly tested in a series of pilots and constantly
optimized. Feedback is then collected and implemented as best as possible
before finally being validated with 3 families having children of various ages
and a variety of responses. The product performed quite well in terms of capture
speed though the mesh representation left something to be desired.

Head
Scanner
3d scanning
Photogrammetry
Down syndrome

http://resolver.tudelft.nl/uuid:295e10a0-22df-4a56-a4c4-351fac597fdd

Student theses

master thesis

© 2021 Yaman Gupta