Translator Disclaimer
9 August 2018 Stereo matching based 3D model reconstruction
Author Affiliations +
Proceedings Volume 10806, Tenth International Conference on Digital Image Processing (ICDIP 2018); 108063A (2018) https://doi.org/10.1117/12.2502843
Event: Tenth International Conference on Digital Image Processing (ICDIP 2018), 2018, Shanghai, China
Abstract
To acquire the precise 3D space information, this paper proposed a stereo matching method based on key region to obtain 3D models. Firstly, we extract matched feature points by creating nonlinear scale spaces with efficient local feature descriptors. Subsequently, object areas are localized based on these feature points using an improved region growing algorithm. Finally, we utilize an improved stereo matching method with three different pixel features matching costs and the non-local cost aggregation to compute the disparity map. The original 3D position information of the corresponding area can be computed by the map. The accuracy and efficiency of the proposed method are evaluated quantitatively on benchmark data.
© (2018) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Lijie Yuan, Yun Tie, and Lin Qi "Stereo matching based 3D model reconstruction", Proc. SPIE 10806, Tenth International Conference on Digital Image Processing (ICDIP 2018), 108063A (9 August 2018); https://doi.org/10.1117/12.2502843
PROCEEDINGS
7 PAGES


SHARE
Advertisement
Advertisement
RELATED CONTENT

3D reconstruction of surface based on binocular vision
Proceedings of SPIE (October 10 2013)
3D reconstruction based on light field images
Proceedings of SPIE (April 10 2018)
Computer vision reconstructing 3 D model from 2 D...
Proceedings of SPIE (March 01 1992)
Three-dimensional object surface matching
Proceedings of SPIE (June 16 1995)

Back to Top