View priority based threads allocation and binary search oriented reweight for GPU accelerated real-time 3D ball tracking

Yilin Hou, Ziwei Deng, Xina Cheng, Takeshi Ikenaga

Research output: Contribution to journalArticle

Abstract

In real-time 3D ball tracking of sports analysis in computer vision technology, complex algorithms which assure the accuracy could be time-consuming. Particle filter based algorithm has a large potential to accelerate since the algorithm between particles has the chance to be paralleled in heterogeneous CPU-GPU platform. Still, with the target multi-view 3D ball tracking algorithm, challenges exist: 1) serial flowchart for each step in the algorithm; 2) repeated processing for multiple views' processing; 3) the low degree of parallelism in reweight and resampling steps for sequential processing. On the CPU-GPU platform, this paper proposes the double stream system flow, the view priority based threads allocation, and the binary search oriented reweight. Double stream system flow assigns tasks which there is no data dependency exists into different streams for each frame processing to achieve parallelism in system structure level. View priority based threads allocation manipulates threads in multi-view observation task. Threads number is view number multiplied by particles number, and with view priority assigning, which could help both memory accessing and computing achieving parallelism. Binary search oriented reweight reduces the time complexity by avoiding to generate cumulative distribution function and uses an unordered array to implement a binary search. The experiment is based on videos which record the final game of an official volleyball match (2014 Inter-High School Games of Men's Volleyball held in Tokyo Metropolitan Gymnasium in Aug. 2014) and the test sequences are taken by multiple-view system which is made of 4 cameras locating at the four corners of the court. The success rate achieves 99.23% which is the same as target algorithm while the time consumption has been accelerated from 75.1ms/frame in CPU environment to 3.05ms/frame in the proposed system which is 24.62 times speed up, also, it achieves 2.33 times speedup compared with basic GPU implemented work.

Original languageEnglish
Pages (from-to)3190-3198
Number of pages9
JournalIEICE Transactions on Information and Systems
VolumeE101D
Issue number12
DOIs
Publication statusPublished - 2018 Dec 1

Fingerprint

Program processors
Processing
Sports
Computer vision
Distribution functions
Graphics processing unit
Cameras
Data storage equipment
Experiments

Keywords

  • Ball tracking
  • Graphics Processing Unit (GPU)
  • Parallelism
  • Particle filter
  • Real-time
  • Thread allocation

ASJC Scopus subject areas

  • Software
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering
  • Artificial Intelligence

Cite this

View priority based threads allocation and binary search oriented reweight for GPU accelerated real-time 3D ball tracking. / Hou, Yilin; Deng, Ziwei; Cheng, Xina; Ikenaga, Takeshi.

In: IEICE Transactions on Information and Systems, Vol. E101D, No. 12, 01.12.2018, p. 3190-3198.

Research output: Contribution to journalArticle

@article{a18a1bc5151b486c93c5b161aad84679,
title = "View priority based threads allocation and binary search oriented reweight for GPU accelerated real-time 3D ball tracking",
abstract = "In real-time 3D ball tracking of sports analysis in computer vision technology, complex algorithms which assure the accuracy could be time-consuming. Particle filter based algorithm has a large potential to accelerate since the algorithm between particles has the chance to be paralleled in heterogeneous CPU-GPU platform. Still, with the target multi-view 3D ball tracking algorithm, challenges exist: 1) serial flowchart for each step in the algorithm; 2) repeated processing for multiple views' processing; 3) the low degree of parallelism in reweight and resampling steps for sequential processing. On the CPU-GPU platform, this paper proposes the double stream system flow, the view priority based threads allocation, and the binary search oriented reweight. Double stream system flow assigns tasks which there is no data dependency exists into different streams for each frame processing to achieve parallelism in system structure level. View priority based threads allocation manipulates threads in multi-view observation task. Threads number is view number multiplied by particles number, and with view priority assigning, which could help both memory accessing and computing achieving parallelism. Binary search oriented reweight reduces the time complexity by avoiding to generate cumulative distribution function and uses an unordered array to implement a binary search. The experiment is based on videos which record the final game of an official volleyball match (2014 Inter-High School Games of Men's Volleyball held in Tokyo Metropolitan Gymnasium in Aug. 2014) and the test sequences are taken by multiple-view system which is made of 4 cameras locating at the four corners of the court. The success rate achieves 99.23{\%} which is the same as target algorithm while the time consumption has been accelerated from 75.1ms/frame in CPU environment to 3.05ms/frame in the proposed system which is 24.62 times speed up, also, it achieves 2.33 times speedup compared with basic GPU implemented work.",
keywords = "Ball tracking, Graphics Processing Unit (GPU), Parallelism, Particle filter, Real-time, Thread allocation",
author = "Yilin Hou and Ziwei Deng and Xina Cheng and Takeshi Ikenaga",
year = "2018",
month = "12",
day = "1",
doi = "10.1587/transinf.2018EDP7125",
language = "English",
volume = "E101D",
pages = "3190--3198",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "12",

}

TY - JOUR

T1 - View priority based threads allocation and binary search oriented reweight for GPU accelerated real-time 3D ball tracking

AU - Hou, Yilin

AU - Deng, Ziwei

AU - Cheng, Xina

AU - Ikenaga, Takeshi

PY - 2018/12/1

Y1 - 2018/12/1

N2 - In real-time 3D ball tracking of sports analysis in computer vision technology, complex algorithms which assure the accuracy could be time-consuming. Particle filter based algorithm has a large potential to accelerate since the algorithm between particles has the chance to be paralleled in heterogeneous CPU-GPU platform. Still, with the target multi-view 3D ball tracking algorithm, challenges exist: 1) serial flowchart for each step in the algorithm; 2) repeated processing for multiple views' processing; 3) the low degree of parallelism in reweight and resampling steps for sequential processing. On the CPU-GPU platform, this paper proposes the double stream system flow, the view priority based threads allocation, and the binary search oriented reweight. Double stream system flow assigns tasks which there is no data dependency exists into different streams for each frame processing to achieve parallelism in system structure level. View priority based threads allocation manipulates threads in multi-view observation task. Threads number is view number multiplied by particles number, and with view priority assigning, which could help both memory accessing and computing achieving parallelism. Binary search oriented reweight reduces the time complexity by avoiding to generate cumulative distribution function and uses an unordered array to implement a binary search. The experiment is based on videos which record the final game of an official volleyball match (2014 Inter-High School Games of Men's Volleyball held in Tokyo Metropolitan Gymnasium in Aug. 2014) and the test sequences are taken by multiple-view system which is made of 4 cameras locating at the four corners of the court. The success rate achieves 99.23% which is the same as target algorithm while the time consumption has been accelerated from 75.1ms/frame in CPU environment to 3.05ms/frame in the proposed system which is 24.62 times speed up, also, it achieves 2.33 times speedup compared with basic GPU implemented work.

AB - In real-time 3D ball tracking of sports analysis in computer vision technology, complex algorithms which assure the accuracy could be time-consuming. Particle filter based algorithm has a large potential to accelerate since the algorithm between particles has the chance to be paralleled in heterogeneous CPU-GPU platform. Still, with the target multi-view 3D ball tracking algorithm, challenges exist: 1) serial flowchart for each step in the algorithm; 2) repeated processing for multiple views' processing; 3) the low degree of parallelism in reweight and resampling steps for sequential processing. On the CPU-GPU platform, this paper proposes the double stream system flow, the view priority based threads allocation, and the binary search oriented reweight. Double stream system flow assigns tasks which there is no data dependency exists into different streams for each frame processing to achieve parallelism in system structure level. View priority based threads allocation manipulates threads in multi-view observation task. Threads number is view number multiplied by particles number, and with view priority assigning, which could help both memory accessing and computing achieving parallelism. Binary search oriented reweight reduces the time complexity by avoiding to generate cumulative distribution function and uses an unordered array to implement a binary search. The experiment is based on videos which record the final game of an official volleyball match (2014 Inter-High School Games of Men's Volleyball held in Tokyo Metropolitan Gymnasium in Aug. 2014) and the test sequences are taken by multiple-view system which is made of 4 cameras locating at the four corners of the court. The success rate achieves 99.23% which is the same as target algorithm while the time consumption has been accelerated from 75.1ms/frame in CPU environment to 3.05ms/frame in the proposed system which is 24.62 times speed up, also, it achieves 2.33 times speedup compared with basic GPU implemented work.

KW - Ball tracking

KW - Graphics Processing Unit (GPU)

KW - Parallelism

KW - Particle filter

KW - Real-time

KW - Thread allocation

UR - http://www.scopus.com/inward/record.url?scp=85057528826&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85057528826&partnerID=8YFLogxK

U2 - 10.1587/transinf.2018EDP7125

DO - 10.1587/transinf.2018EDP7125

M3 - Article

VL - E101D

SP - 3190

EP - 3198

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 12

ER -