Acceleration of the transformation from elliptic omnidirectional images to panoramic images using graphic processing units

Cheng Hung Lin; Wen Jui Chou

doi:10.1109/ICCE-TW.2016.7520975

Acceleration of the transformation from elliptic omnidirectional images to panoramic images using graphic processing units

Cheng Hung Lin, Wen Jui Chou

Department of Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Citation (Scopus)

Abstract

Omni-directional cameras are widely used in many applications such as surveillance systems and endoscopy. Omnidirectional cameras use a single camera and a reflective mirror to capture elliptic omnidirectional images and then transform the elliptic omnidirectional images to panoramic images. To accelerate the transformation from elliptic omnidirectional images to panoramic images, this paper proposes a hierarchical parallelism including data parallelism and task parallelism to improve the performance of transformation using graphic processing units. The data parallelism accelerates the mapping of pixels from elliptic omnidirectional images to panoramic images using multiple threads simultaneously while the task parallelism performs deep pipelines on multiple streams. We have implemented the proposed algorithm using CUDA on NVIDIA GPUs. The experimental results show that the proposed hierarchical parallelism performed on GPUs achieves 6.33 times faster than the CPU counterpart does.

Original language	English
Title of host publication	2016 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781509020737
DOIs	https://doi.org/10.1109/ICCE-TW.2016.7520975
Publication status	Published - 2016 Jul 25
Event	3rd IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016 - Nantou County, Taiwan Duration: 2016 May 27 → 2016 May 30

Publication series

Name	2016 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016

Other

Other	3rd IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016
Country/Territory	Taiwan
City	Nantou County
Period	2016/05/27 → 2016/05/30

Keywords

Omni-directional camera
graphic processing units
panoramic image
parallelism

ASJC Scopus subject areas

Computer Science Applications
Electrical and Electronic Engineering
Media Technology
Computer Networks and Communications
Signal Processing

Access to Document

10.1109/ICCE-TW.2016.7520975

Cite this

Lin, C. H., & Chou, W. J. (2016). Acceleration of the transformation from elliptic omnidirectional images to panoramic images using graphic processing units. In 2016 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016 Article 7520975 (2016 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCE-TW.2016.7520975

Acceleration of the transformation from elliptic omnidirectional images to panoramic images using graphic processing units. / Lin, Cheng Hung; Chou, Wen Jui.
2016 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016. Institute of Electrical and Electronics Engineers Inc., 2016. 7520975 (2016 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Lin, CH & Chou, WJ 2016, Acceleration of the transformation from elliptic omnidirectional images to panoramic images using graphic processing units. in 2016 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016., 7520975, 2016 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016, Institute of Electrical and Electronics Engineers Inc., 3rd IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016, Nantou County, Taiwan, 2016/05/27. https://doi.org/10.1109/ICCE-TW.2016.7520975

Lin CH, Chou WJ. Acceleration of the transformation from elliptic omnidirectional images to panoramic images using graphic processing units. In 2016 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016. Institute of Electrical and Electronics Engineers Inc. 2016. 7520975. (2016 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016). doi: 10.1109/ICCE-TW.2016.7520975

Lin, Cheng Hung ; Chou, Wen Jui. / Acceleration of the transformation from elliptic omnidirectional images to panoramic images using graphic processing units. 2016 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016. Institute of Electrical and Electronics Engineers Inc., 2016. (2016 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016).

@inproceedings{ab08d1e9f87145bb91c11af0b5d0e4f5,

title = "Acceleration of the transformation from elliptic omnidirectional images to panoramic images using graphic processing units",

abstract = "Omni-directional cameras are widely used in many applications such as surveillance systems and endoscopy. Omnidirectional cameras use a single camera and a reflective mirror to capture elliptic omnidirectional images and then transform the elliptic omnidirectional images to panoramic images. To accelerate the transformation from elliptic omnidirectional images to panoramic images, this paper proposes a hierarchical parallelism including data parallelism and task parallelism to improve the performance of transformation using graphic processing units. The data parallelism accelerates the mapping of pixels from elliptic omnidirectional images to panoramic images using multiple threads simultaneously while the task parallelism performs deep pipelines on multiple streams. We have implemented the proposed algorithm using CUDA on NVIDIA GPUs. The experimental results show that the proposed hierarchical parallelism performed on GPUs achieves 6.33 times faster than the CPU counterpart does.",

keywords = "Omni-directional camera, graphic processing units, panoramic image, parallelism",

author = "Lin, {Cheng Hung} and Chou, {Wen Jui}",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 3rd IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016 ; Conference date: 27-05-2016 Through 30-05-2016",

year = "2016",

month = jul,

day = "25",

doi = "10.1109/ICCE-TW.2016.7520975",

language = "English",

series = "2016 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2016 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016",

}

TY - GEN

T1 - Acceleration of the transformation from elliptic omnidirectional images to panoramic images using graphic processing units

AU - Lin, Cheng Hung

AU - Chou, Wen Jui

PY - 2016/7/25

Y1 - 2016/7/25

N2 - Omni-directional cameras are widely used in many applications such as surveillance systems and endoscopy. Omnidirectional cameras use a single camera and a reflective mirror to capture elliptic omnidirectional images and then transform the elliptic omnidirectional images to panoramic images. To accelerate the transformation from elliptic omnidirectional images to panoramic images, this paper proposes a hierarchical parallelism including data parallelism and task parallelism to improve the performance of transformation using graphic processing units. The data parallelism accelerates the mapping of pixels from elliptic omnidirectional images to panoramic images using multiple threads simultaneously while the task parallelism performs deep pipelines on multiple streams. We have implemented the proposed algorithm using CUDA on NVIDIA GPUs. The experimental results show that the proposed hierarchical parallelism performed on GPUs achieves 6.33 times faster than the CPU counterpart does.

AB - Omni-directional cameras are widely used in many applications such as surveillance systems and endoscopy. Omnidirectional cameras use a single camera and a reflective mirror to capture elliptic omnidirectional images and then transform the elliptic omnidirectional images to panoramic images. To accelerate the transformation from elliptic omnidirectional images to panoramic images, this paper proposes a hierarchical parallelism including data parallelism and task parallelism to improve the performance of transformation using graphic processing units. The data parallelism accelerates the mapping of pixels from elliptic omnidirectional images to panoramic images using multiple threads simultaneously while the task parallelism performs deep pipelines on multiple streams. We have implemented the proposed algorithm using CUDA on NVIDIA GPUs. The experimental results show that the proposed hierarchical parallelism performed on GPUs achieves 6.33 times faster than the CPU counterpart does.

KW - Omni-directional camera

KW - graphic processing units

KW - panoramic image

KW - parallelism

UR - http://www.scopus.com/inward/record.url?scp=84983504723&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84983504723&partnerID=8YFLogxK

U2 - 10.1109/ICCE-TW.2016.7520975

DO - 10.1109/ICCE-TW.2016.7520975

M3 - Conference contribution

AN - SCOPUS:84983504723

T3 - 2016 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016

BT - 2016 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 3rd IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2016

Y2 - 27 May 2016 through 30 May 2016

ER -

Acceleration of the transformation from elliptic omnidirectional images to panoramic images using graphic processing units

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this