Abstract
Single-shot structured light profilometry (SLP) aims at reconstructing the 3D height map of an object from a single deformed fringe pattern and has long been the ultimate goal in fringe projection profilometry. Recently, deep learning was introduced into SLP setups to replace the task-specific algorithm of fringe demodulation with a dedicated neural network. Research on deep learning-based profilometry has made considerable progress in a short amount of time due to the rapid development of general neural network strategies and to the transferrable nature of deep learning techniques to a wide array of application fields. The selection of the employed loss function has received very little to no attention in the recently reported deep learning-based SLP setups. In this paper, we demonstrate the significant impact of loss function selection on height map prediction accuracy, we evaluate the performance of a range of commonly used loss functions and we propose a new mixed gradient loss function that yields a higher 3D surface reconstruction accuracy than any previously used loss functions.
Highlights
Structured light profilometry (SLP) is the technique of reconstructing the 3D surface map of an object by projecting predefined fringe patterns onto its surface and by observing it under an angle [1, 2]
We demonstrate the significant impact of loss function selection on height map prediction accuracy, we evaluate the performance of a range of commonly used loss functions and we propose a new mixed gradient loss function that yields a higher 3D surface reconstruction accuracy than any previously used loss functions
We demonstrate that the choice of loss function in a deep learning-based SLP setup has a significant impact on prediction accuracy, we evaluate the performance of several common loss functions and we propose a custom mixed gradient loss function that yields a higher prediction accuracy than any of the other investigated loss functions
Summary
Structured light profilometry (SLP) is the technique of reconstructing the 3D surface map of an object by projecting predefined fringe patterns onto its surface and by observing it under an angle [1, 2]. From the recorded deformed fringe pattern, the full-field height map of the object can be extracted using various triangulation techniques that can be subdivided into different classes depending on the specific demodulation strategy that is employed. All single-shot strategies are either adapted versions of Takeda’s Fourier transform profilometry (FTP) [5] or employ additional color channels to superimpose differently modulated intensity patterns onto a single color map [6, 7]. While both approaches have been highly successful and are implemented into a variety of application fields, they each have specific drawbacks. When this is not the case, the quality of the resulting depth map deteriorates significantly
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
