3D Motion Capture on Graphic Processing Unit using CUDA

อภิสิทธิ์ รัตนาตรานุรักษ์, ธนพนธ์ จันทรวงศ์สาลี, ไอราวัณ มงคลการ, ศิริชัย จูทิม

Abstract


This paper presents 3D Motion Capture on Graphic Processing Unit using CUDA which is the framework for developing program on NVIDIA Graphic Processing Unit. The paper starts with analysis run time to construct each part of model such as right arm and right hand, left arm and left hand, head and body, right leg and right foot, left leg and left foot. We then implemented CUDA to this algorithm to execute the model construction algorithm on Graphic Processing Unit. The results show that average Speed up of Graphic Processing Unit is 73.74x on 96 CUDA cores.


Keywords


Graphic Processing Unit, 3D Motion Capture, CUDA.

Full Text:

PDF 25-36

References


AMD (2011). OpenCL programming

guide. จาก http://developer.amd.

com/sdks/AMDAPPSDK/docum

entation.

Bian, Y., Zhao, X., Song, J., & Liu, Y.

(2012, November). Parallelized

annealed particle filter for

real-time marker-less motion

tracking via heterogeneous computing.

In Pattern Recognition (ICPR), 2012

st International Conference

on (pp. 2444-2447). IEEE.

Khronos OpenCL Working Group

(2010). The OpenCL Specification

(version 1.1).

Kwolek, B., Krzeszowski, T., &

Wojciechowski, K. (2011, August).

Swarm intelligence based

searching schemes for articulated 3D

body motion tracking. In International

Conference on Advanced

Concepts for Intelligent Vision

Systems (pp. 115-126). Springer

Berlin Heidelberg.

Moeslund, T. B., & Granum, E. (2001).

A survey of computer visionbased

human motion capture.

Computer vision and image

understanding, 81(3), 231-268.

Mokhov, S. A., Song, M., Llewellyn,

J., Zhang, J., Charette, A., Wu,

R., & Ge, S. (2016, July). Realtime

collection and analysis of

-Kinect v2 skeleton data in a

single application. In ACM SIGGRAPH

Posters (p. 53). ACM.

Muratov, O., Slynko, Y., Chernov, V.,

Lyubimtseva, M., Shamsuarov,

A., & Bucha, V. (2016).

DCapture: 3D Reconstruction

for a Smartphone. In Proceedings of

the IEEE Conference on Computer

Vision and Pattern Recognition

Workshops (pp. 75-82).

NVIDIA, C. U. D. A. (2007). Compute

unified device architecture

programming guide.

NVIDIA. (2009). NVIDIA OpenCL Jump

start guide

Shujun, Z., Cong, W., Xuqiang, S., &

Wei, W. (2009, May). DreamWorld:

CUDA-accelerated real-time 3D

modeling system. In 2009 IEEE

International Conference on

Virtual Environments, Human-

Computer Interfaces and Measurements

Systems (pp. 168-173). IEEE.

Wei, X., Zhang, P., & Chai, J. (2012).

Accurate realtime full-body

motion capture using a single

depth camera. ACM Transactions on

Graphics (TOG), 31(6), 188.

Zhang, Z., & Seah, H. S. (2012, December).

CUDA acceleration of 3D dynamic

scene reconstruction and 3D

motion estimation for motion

capture. In Parallel and Distributed

Systems (ICPADS), 2012 IEEE

th International Conference

on (pp. 284-291). IEEE.


Refbacks

  • There are currently no refbacks.


Copyright (c) 2017 วารสารวิชาการเทคโนโลยีอุตสาหกรรม : มหาวิทยาลัยราชภัฏสวนสุนันทา The Journal of Industrial Technology

Faculty of Industrial Technology Suan Sunandha Rajabhat University 1 U-tongnok Dusit Bangkok 10300  Tel. 66 2160 1438#22  E-mail. fit@ssru.ac.th