Skip navigation

Physical Scans

Authors: Sebastian Koch
Yurii Piadyk
Markus Worchel
Marc Alexa
Claudio Silva
Denis Zorin
Daniele Panozzo
Issue Date: 2021
Abstract: Images of a real scene taken with a camera commonly differ from synthetic images of a virtual replica of the same scene, despite advances in light transport simulation and calibration. By explicitly co-developing the scanning hardware and rendering pipeline we are able to achieve negligible per-pixel difference between the real image taken by the camera and the synthesized image on geometrically complex calibration object with known material properties. This approach provides an ideal test-bed for developing data-driven algorithms in the area of 3D reconstruction, as the synthetic data is indistinguishable from real data and can be generated at large scale. Pixel-wise matching also provides an effective way to quantitatively evaluate data-driven reconstruction algorithms.
Description: Physical scans of 3 (Pawn, Rook and Shapes) machined white calibration objects and 7 (Dodo, Vessel/Bird, House, Radio, Sculpture/Avocado, Chair, Vase) 3D printed color textured test objects. Calibration objects have also been scanned with ambient lights off, including an additional scan of the flat plane object for material calibration. Rotating stage calibration was performed before and after scanning the objects for cross-checking for any disturbances. Additional stage calibration was performed for a pair of Pawn scans with matte and glossy material coating. Collections of patterns used are also included.
Rights: CC BY 4.0 License
Appears in Collections:Hardware Design and Accurate Simulation for Benchmarking of 3D Reconstruction Algorithms

Items in FDA are protected by copyright, with all rights reserved, unless otherwise indicated.