This work introduces a novel optical 3D sensor principle and its implementation: the ‘single-shot 3D movie camera’. The camera is designed for the 3D measurement of macroscopic objects with scattering surfaces, e.g. human faces. It combines the acquisition of a dense point cloud displaying physically limited lateral resolution and depth precision together with a single-shot ability. ‘Single-shot’ means that no temporal sequence of exposures is exploited to generate the 3D point cloud. The approach is based on multi-line triangulation. Since, in contrast to other single-shot approaches, no space-bandwidth is wasted by pattern codification, the 3D point cloud can be acquired with its maximal possible density: A 1-Megapixel camera (1000 × 1000 pix) delivers nearly 300,000 independent (uncorrelated) 3D points in each camera frame. A 3D sensor with these features allows for a continuous 3D measurement of moving or deforming objects, resulting in a ‘3D movie’.

Last modified

• 06/04/2018

Creator

• Florian Willomitzer

DOI

• https://doi.org/10.21985/N2N98J

Keyword

• crd18

Date created

• 2018-04-10

Resource type

• Video

Rights statement

• In Copyright