G. Payen de La Garanderie
Improved Depth Recovery In Consumer Depth Cameras via Disparity Space Fusion within Cross-spectral Stereo
Payen de La Garanderie, G.; Breckon, T.P.
Authors
Professor Toby Breckon toby.breckon@durham.ac.uk
Summer School Academic (Casual)
Contributors
Michel Valstar
Editor
Andrew French
Editor
Tony Pridmore
Editor
Abstract
We address the issue of improving depth coverage in consumer depth cameras based on the combined use of cross-spectral stereo and near infra-red structured light sensing. Specifically we show that fusion of disparity over these modalities, within the disparity space image, prior to disparity optimization facilitates the recovery of scene depth information in regions where structured light sensing fails. We show that this joint approach, leveraging disparity information from both structured light and cross-spectral sensing, facilitates the joint recovery of global scene depth comprising both texture-less object depth, where conventional stereo otherwise fails, and highly reflective object depth, where structured light (and similar) active sensing commonly fails. The proposed solution is illustrated using dense gradient feature matching and shown to outperform prior approaches that use late-stage fused cross-spectral stereo depth as a facet of improved sensing for consumer depth cameras.
Citation
Payen de La Garanderie, G., & Breckon, T. (2014). Improved Depth Recovery In Consumer Depth Cameras via Disparity Space Fusion within Cross-spectral Stereo. In M. Valstar, A. French, & T. Pridmore (Eds.), Proceedings of the British Machine Vision Conference (417.1-417.12). https://doi.org/10.5244/C.28.110
Presentation Conference Type | Conference Paper (Published) |
---|---|
Conference Name | Proceedings of the British Machine Vision Conference |
Publication Date | Sep 1, 2014 |
Deposit Date | Dec 9, 2014 |
Publicly Available Date | Feb 4, 2015 |
Pages | 417.1-417.12 |
Book Title | Proceedings of the British Machine Vision Conference. |
DOI | https://doi.org/10.5244/C.28.110 |
Keywords | rgb-d, consumer depth camera, kinect, depth improvment, disparity space image, multimodal stereo, multi-modal, infrared colour stereo, depth map improvement, hole filling, structured light, disparity filling, disparity improvement, infrared stereo vision |
Public URL | https://durham-repository.worktribe.com/output/1154173 |
Publisher URL | https://web.archive.org/web/20220706011347/https://www.bmva.org/bmvc/2014/papers/paper107/index.html |
Related Public URLs | https://breckon.org/toby/publications/papers/pdlg14depth.pdf |
Files
Published Conference Proceeding
(4.5 Mb)
PDF
You might also like
Back to Butterworth - a Fourier Basis for 3D Surface Relief Hole Filling within RGB-D Imagery
(2016)
Presentation / Conference Contribution
RAPiD-Seg: Range-Aware Pointwise Distance Distribution Networks for LiDAR Semantic Segmentation
(2024)
Presentation / Conference Contribution
Towards Open-World Object-based Anomaly Detection via Self-Supervised Outlier Synthesis
(2024)
Presentation / Conference Contribution
U3DS3 : Unsupervised 3D Semantic Scene Segmentation
(2024)
Presentation / Conference Contribution
Downloadable Citations
About Durham Research Online (DRO)
Administrator e-mail: dro.admin@durham.ac.uk
This application uses the following open-source libraries:
SheetJS Community Edition
Apache License Version 2.0 (http://www.apache.org/licenses/)
PDF.js
Apache License Version 2.0 (http://www.apache.org/licenses/)
Font Awesome
SIL OFL 1.1 (http://scripts.sil.org/OFL)
MIT License (http://opensource.org/licenses/mit-license.html)
CC BY 3.0 ( http://creativecommons.org/licenses/by/3.0/)
Powered by Worktribe © 2024
Advanced Search