Real-time 3D ball tracking with CPU-GPU acceleration using particle filter with multi-command queues and stepped parallelism iteration

Yilin Hou, Xina Cheng, Takeshi Ikenaga

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

7 Citations (Scopus)

Abstract

3D ball tracking is a critical function in manyapplications such as game and players behavior analysis, andreal time implementation has become increasingly importantfor it can be used for live broadcast and TV contents. To reacha high accuracy, algorithms usually are time consuming due toa large set of calculations which is challenging to meet realtime demanding. This paper proposes multiple commandqueues, tactical threads allocation and stepped iterativeaddition to empower such a capacity on the CPU-GPUplatform. Multiple command queues achieves a parallelismbetween tasks in the algorithm. Secondly, the tactical threadsallocation helps mapping the algorithm into GPU andenhances synchronism between threads. And this paperproposes stepped iterative addition to achieve partialparallelism in a sequential operation. This work implements inan Intel Core i7-6700 GPU and AMD Radeon R9 FURY GPU.Tracking speed of our work increases 37.8 times from original431ms to 11.7ms while the success rate of the algorithm retainsover 99%. This result fully meets the requirement of 16.6msper frame for 60fps video real-time tracking.

Original languageEnglish
Title of host publicationProceedings - 2017 2nd International Conference on Multimedia and Image Processing, ICMIP 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages235-239
Number of pages5
Volume2017-January
ISBN (Electronic)9781509059546
DOIs
Publication statusPublished - 2017 Dec 15
Event2nd International Conference on Multimedia and Image Processing, ICMIP 2017 - Wuhan, Hubei, China
Duration: 2017 Mar 172017 Mar 19

Other

Other2nd International Conference on Multimedia and Image Processing, ICMIP 2017
CountryChina
CityWuhan, Hubei
Period17/3/1717/3/19

Fingerprint

Program processors
Synchronization
Graphics processing unit

Keywords

  • GPU
  • Hardware acceleration
  • Particle filter
  • Sports analysis

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Media Technology

Cite this

Hou, Y., Cheng, X., & Ikenaga, T. (2017). Real-time 3D ball tracking with CPU-GPU acceleration using particle filter with multi-command queues and stepped parallelism iteration. In Proceedings - 2017 2nd International Conference on Multimedia and Image Processing, ICMIP 2017 (Vol. 2017-January, pp. 235-239). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICMIP.2017.59

Real-time 3D ball tracking with CPU-GPU acceleration using particle filter with multi-command queues and stepped parallelism iteration. / Hou, Yilin; Cheng, Xina; Ikenaga, Takeshi.

Proceedings - 2017 2nd International Conference on Multimedia and Image Processing, ICMIP 2017. Vol. 2017-January Institute of Electrical and Electronics Engineers Inc., 2017. p. 235-239.

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

Hou, Y, Cheng, X & Ikenaga, T 2017, Real-time 3D ball tracking with CPU-GPU acceleration using particle filter with multi-command queues and stepped parallelism iteration. in Proceedings - 2017 2nd International Conference on Multimedia and Image Processing, ICMIP 2017. vol. 2017-January, Institute of Electrical and Electronics Engineers Inc., pp. 235-239, 2nd International Conference on Multimedia and Image Processing, ICMIP 2017, Wuhan, Hubei, China, 17/3/17. https://doi.org/10.1109/ICMIP.2017.59
Hou Y, Cheng X, Ikenaga T. Real-time 3D ball tracking with CPU-GPU acceleration using particle filter with multi-command queues and stepped parallelism iteration. In Proceedings - 2017 2nd International Conference on Multimedia and Image Processing, ICMIP 2017. Vol. 2017-January. Institute of Electrical and Electronics Engineers Inc. 2017. p. 235-239 https://doi.org/10.1109/ICMIP.2017.59
Hou, Yilin ; Cheng, Xina ; Ikenaga, Takeshi. / Real-time 3D ball tracking with CPU-GPU acceleration using particle filter with multi-command queues and stepped parallelism iteration. Proceedings - 2017 2nd International Conference on Multimedia and Image Processing, ICMIP 2017. Vol. 2017-January Institute of Electrical and Electronics Engineers Inc., 2017. pp. 235-239
@inproceedings{f3bc6e8d8ef24839a4db341722442935,
title = "Real-time 3D ball tracking with CPU-GPU acceleration using particle filter with multi-command queues and stepped parallelism iteration",
abstract = "3D ball tracking is a critical function in manyapplications such as game and players behavior analysis, andreal time implementation has become increasingly importantfor it can be used for live broadcast and TV contents. To reacha high accuracy, algorithms usually are time consuming due toa large set of calculations which is challenging to meet realtime demanding. This paper proposes multiple commandqueues, tactical threads allocation and stepped iterativeaddition to empower such a capacity on the CPU-GPUplatform. Multiple command queues achieves a parallelismbetween tasks in the algorithm. Secondly, the tactical threadsallocation helps mapping the algorithm into GPU andenhances synchronism between threads. And this paperproposes stepped iterative addition to achieve partialparallelism in a sequential operation. This work implements inan Intel Core i7-6700 GPU and AMD Radeon R9 FURY GPU.Tracking speed of our work increases 37.8 times from original431ms to 11.7ms while the success rate of the algorithm retainsover 99{\%}. This result fully meets the requirement of 16.6msper frame for 60fps video real-time tracking.",
keywords = "GPU, Hardware acceleration, Particle filter, Sports analysis",
author = "Yilin Hou and Xina Cheng and Takeshi Ikenaga",
year = "2017",
month = "12",
day = "15",
doi = "10.1109/ICMIP.2017.59",
language = "English",
volume = "2017-January",
pages = "235--239",
booktitle = "Proceedings - 2017 2nd International Conference on Multimedia and Image Processing, ICMIP 2017",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Real-time 3D ball tracking with CPU-GPU acceleration using particle filter with multi-command queues and stepped parallelism iteration

AU - Hou, Yilin

AU - Cheng, Xina

AU - Ikenaga, Takeshi

PY - 2017/12/15

Y1 - 2017/12/15

N2 - 3D ball tracking is a critical function in manyapplications such as game and players behavior analysis, andreal time implementation has become increasingly importantfor it can be used for live broadcast and TV contents. To reacha high accuracy, algorithms usually are time consuming due toa large set of calculations which is challenging to meet realtime demanding. This paper proposes multiple commandqueues, tactical threads allocation and stepped iterativeaddition to empower such a capacity on the CPU-GPUplatform. Multiple command queues achieves a parallelismbetween tasks in the algorithm. Secondly, the tactical threadsallocation helps mapping the algorithm into GPU andenhances synchronism between threads. And this paperproposes stepped iterative addition to achieve partialparallelism in a sequential operation. This work implements inan Intel Core i7-6700 GPU and AMD Radeon R9 FURY GPU.Tracking speed of our work increases 37.8 times from original431ms to 11.7ms while the success rate of the algorithm retainsover 99%. This result fully meets the requirement of 16.6msper frame for 60fps video real-time tracking.

AB - 3D ball tracking is a critical function in manyapplications such as game and players behavior analysis, andreal time implementation has become increasingly importantfor it can be used for live broadcast and TV contents. To reacha high accuracy, algorithms usually are time consuming due toa large set of calculations which is challenging to meet realtime demanding. This paper proposes multiple commandqueues, tactical threads allocation and stepped iterativeaddition to empower such a capacity on the CPU-GPUplatform. Multiple command queues achieves a parallelismbetween tasks in the algorithm. Secondly, the tactical threadsallocation helps mapping the algorithm into GPU andenhances synchronism between threads. And this paperproposes stepped iterative addition to achieve partialparallelism in a sequential operation. This work implements inan Intel Core i7-6700 GPU and AMD Radeon R9 FURY GPU.Tracking speed of our work increases 37.8 times from original431ms to 11.7ms while the success rate of the algorithm retainsover 99%. This result fully meets the requirement of 16.6msper frame for 60fps video real-time tracking.

KW - GPU

KW - Hardware acceleration

KW - Particle filter

KW - Sports analysis

UR - http://www.scopus.com/inward/record.url?scp=85033442376&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85033442376&partnerID=8YFLogxK

U2 - 10.1109/ICMIP.2017.59

DO - 10.1109/ICMIP.2017.59

M3 - Conference contribution

VL - 2017-January

SP - 235

EP - 239

BT - Proceedings - 2017 2nd International Conference on Multimedia and Image Processing, ICMIP 2017

PB - Institute of Electrical and Electronics Engineers Inc.

ER -