Single-wavelength and multi-parallel dotted- A nd solid-lines for dense and robust active 3D reconstruction

Genki Nagamatsu, Ryo Furukawa, Ryusuke Sagawa, Hiroshi Kawasaki

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A dense one-shot scanning technique that is robust to subsurface scattering is proposed. In this technique, a novel pattern, consisting of multiple parallel dotted lines and solid lines, that are aligned alternately, is proposed. To project such a pattern efficiently, a single-wavelength laser-based pattern projector is developed. To detect patterns robustly from captured images, a black and white camera attached with a narrow-band-path filter is used in conjunction with our novel deep learning based algorithm, which is based on a convolutional neural network (CNN). Because the detected lines must be identified for shape reconstruction, we apply a gap-coding technique, which is originally based on a grid-line pattern, to the dot pattern. To this end, we introduce a virtual grid-line structure, which is generated from the dot pattern. Additionally, we propose a calibration algorithm specialized for our system, where the pattern is static and shared with the shape reconstruction algorithm, i. e., correspondence problem remains. For a solution, gap-coding is further applied to find correspondences under epipolar constraints. The experimental results of scanning real objects are presented to demonstrate the effectiveness of our calibration and reconstruction techniques.

Original languageEnglish
Title of host publicationProceedings of the 16th International Conference on Machine Vision Applications, MVA 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9784901122184
DOIs
Publication statusPublished - May 2019
Event16th International Conference on Machine Vision Applications, MVA 2019 - Tokyo, Japan
Duration: May 27 2019May 31 2019

Publication series

NameProceedings of the 16th International Conference on Machine Vision Applications, MVA 2019

Conference

Conference16th International Conference on Machine Vision Applications, MVA 2019
CountryJapan
CityTokyo
Period5/27/195/31/19

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Signal Processing
  • Computer Vision and Pattern Recognition

Fingerprint Dive into the research topics of 'Single-wavelength and multi-parallel dotted- A nd solid-lines for dense and robust active 3D reconstruction'. Together they form a unique fingerprint.

  • Cite this

    Nagamatsu, G., Furukawa, R., Sagawa, R., & Kawasaki, H. (2019). Single-wavelength and multi-parallel dotted- A nd solid-lines for dense and robust active 3D reconstruction. In Proceedings of the 16th International Conference on Machine Vision Applications, MVA 2019 [8758011] (Proceedings of the 16th International Conference on Machine Vision Applications, MVA 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.23919/MVA.2019.8758011