Kinect-based easy 3D object reconstruction

Di Xu; Jianfei Cai; Tat Jen Cham; Philip Fu; Juyong Zhang

doi:10.1007/978-3-642-34778-8_44

Kinect-based easy 3D object reconstruction

Di Xu
, Jianfei Cai
, Tat Jen Cham
, Philip Fu
, Juyong Zhang

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

7 Citations (Scopus)

Abstract

Inspired by the recently developed KinectFusion technique, which is able to reconstruct a 3D scene in real time through moving Kinect, we consider improving KinectFusion for 3D reconstruction of a real object. We make some adaptations to KinectFusion so as to identify the object-of-interest and separate the 3D object model from the entire 3D scene. Moreover, considering that the 3D object model generated by KinectFusion often contains some clearly visible outliers due to the noisy Kinect data, we propose a refinement scheme to remove the outliers. Our basic idea is to make use of the existing powerful 2D segmentation tool to refine the silhouette in each color image and then form visual hull via the refined dense silhouettes to improve the 3D object model. Experimental results show improved performance.

Original language	English
Title of host publication	Advances in Multimedia Information Processing, PCM 2012 - 13th Pacific-Rim Conference on Multimedia, Proceedings
Publisher	Springer
Pages	476-483
Number of pages	8
ISBN (Print)	9783642347771
DOIs	https://doi.org/10.1007/978-3-642-34778-8_44
Publication status	Published - 2012
Externally published	Yes
Event	Pacific-Rim Conference on Multimedia 2012 - Singapore, Singapore Duration: 4 Dec 2012 → 6 Dec 2012 Conference number: 13th https://link.springer.com/book/10.1007/978-3-642-34778-8 (Conference Proceedings)

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	7674 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	Pacific-Rim Conference on Multimedia 2012
Abbreviated title	PCM 2012
Country/Territory	Singapore
City	Singapore
Period	4/12/12 → 6/12/12
Internet address	https://link.springer.com/book/10.1007/978-3-642-34778-8 (Conference Proceedings)

Keywords

2D segmentation
3D reconstruction
Kinect
visual hull

Access to Document

10.1007/978-3-642-34778-8_44

Cite this

Xu, D., Cai, J., Cham, T. J., Fu, P., & Zhang, J. (2012). Kinect-based easy 3D object reconstruction. In Advances in Multimedia Information Processing, PCM 2012 - 13th Pacific-Rim Conference on Multimedia, Proceedings (pp. 476-483). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7674 LNCS). Springer. https://doi.org/10.1007/978-3-642-34778-8_44

@inproceedings{0a5973a87f1e47f99b607100816c30c8,

title = "Kinect-based easy 3D object reconstruction",

abstract = "Inspired by the recently developed KinectFusion technique, which is able to reconstruct a 3D scene in real time through moving Kinect, we consider improving KinectFusion for 3D reconstruction of a real object. We make some adaptations to KinectFusion so as to identify the object-of-interest and separate the 3D object model from the entire 3D scene. Moreover, considering that the 3D object model generated by KinectFusion often contains some clearly visible outliers due to the noisy Kinect data, we propose a refinement scheme to remove the outliers. Our basic idea is to make use of the existing powerful 2D segmentation tool to refine the silhouette in each color image and then form visual hull via the refined dense silhouettes to improve the 3D object model. Experimental results show improved performance.",

keywords = "2D segmentation, 3D reconstruction, Kinect, visual hull",

author = "Di Xu and Jianfei Cai and Cham, \{Tat Jen\} and Philip Fu and Juyong Zhang",

year = "2012",

doi = "10.1007/978-3-642-34778-8\_44",

language = "English",

isbn = "9783642347771",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer",

pages = "476--483",

booktitle = "Advances in Multimedia Information Processing, PCM 2012 - 13th Pacific-Rim Conference on Multimedia, Proceedings",

address = "Switzerland",

note = "Pacific-Rim Conference on Multimedia 2012, PCM 2012 ; Conference date: 04-12-2012 Through 06-12-2012",

url = "https://link.springer.com/book/10.1007/978-3-642-34778-8",

}

Xu, D, Cai, J, Cham, TJ, Fu, P & Zhang, J 2012, Kinect-based easy 3D object reconstruction. in Advances in Multimedia Information Processing, PCM 2012 - 13th Pacific-Rim Conference on Multimedia, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 7674 LNCS, Springer, pp. 476-483, Pacific-Rim Conference on Multimedia 2012, Singapore, Singapore, 4/12/12. https://doi.org/10.1007/978-3-642-34778-8_44

Kinect-based easy 3D object reconstruction. / Xu, Di; Cai, Jianfei; Cham, Tat Jen et al.
Advances in Multimedia Information Processing, PCM 2012 - 13th Pacific-Rim Conference on Multimedia, Proceedings. Springer, 2012. p. 476-483 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7674 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Kinect-based easy 3D object reconstruction

AU - Xu, Di

AU - Cai, Jianfei

AU - Cham, Tat Jen

AU - Fu, Philip

AU - Zhang, Juyong

N1 - Conference code: 13th

PY - 2012

Y1 - 2012

N2 - Inspired by the recently developed KinectFusion technique, which is able to reconstruct a 3D scene in real time through moving Kinect, we consider improving KinectFusion for 3D reconstruction of a real object. We make some adaptations to KinectFusion so as to identify the object-of-interest and separate the 3D object model from the entire 3D scene. Moreover, considering that the 3D object model generated by KinectFusion often contains some clearly visible outliers due to the noisy Kinect data, we propose a refinement scheme to remove the outliers. Our basic idea is to make use of the existing powerful 2D segmentation tool to refine the silhouette in each color image and then form visual hull via the refined dense silhouettes to improve the 3D object model. Experimental results show improved performance.

AB - Inspired by the recently developed KinectFusion technique, which is able to reconstruct a 3D scene in real time through moving Kinect, we consider improving KinectFusion for 3D reconstruction of a real object. We make some adaptations to KinectFusion so as to identify the object-of-interest and separate the 3D object model from the entire 3D scene. Moreover, considering that the 3D object model generated by KinectFusion often contains some clearly visible outliers due to the noisy Kinect data, we propose a refinement scheme to remove the outliers. Our basic idea is to make use of the existing powerful 2D segmentation tool to refine the silhouette in each color image and then form visual hull via the refined dense silhouettes to improve the 3D object model. Experimental results show improved performance.

KW - 2D segmentation

KW - 3D reconstruction

KW - Kinect

KW - visual hull

UR - https://www.scopus.com/pages/publications/84871454982

U2 - 10.1007/978-3-642-34778-8_44

DO - 10.1007/978-3-642-34778-8_44

M3 - Conference Paper

AN - SCOPUS:84871454982

SN - 9783642347771

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 476

EP - 483

BT - Advances in Multimedia Information Processing, PCM 2012 - 13th Pacific-Rim Conference on Multimedia, Proceedings

PB - Springer

T2 - Pacific-Rim Conference on Multimedia 2012

Y2 - 4 December 2012 through 6 December 2012

ER -

Xu D, Cai J, Cham TJ, Fu P, Zhang J. Kinect-based easy 3D object reconstruction. In Advances in Multimedia Information Processing, PCM 2012 - 13th Pacific-Rim Conference on Multimedia, Proceedings. Springer. 2012. p. 476-483. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-642-34778-8_44

Kinect-based easy 3D object reconstruction

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Cite this