Fast GPU implementation of large scale dictionary and sparse representation based vision problems

Pradeep Nagesh, Rahul Gowda, Baoxin Li

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

8 Citations (Scopus)

Abstract

Recently, Computer Vision problems like Face Recognition and Super-Resolution solved using sparse representation based methods with large dictionaries have shown state-of-the-art results. However such methods are computationally prohibitive for typical CPUs, especially for a large dictionary size. We present fast implementation of these methods by exploiting the massively parallel processing capabilities of a GPU within a CUDA framework, owing to its easy off-the-shelf availability and programmer friendliness. We provide details of system level design, memory management and implementation strategies. Further, we integrate the solution to the preferred scientific computational platform - MATLAB.

Original languageEnglish (US)
Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Pages1570-1573
Number of pages4
DOIs
StatePublished - 2010
Event2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Dallas, TX, United States
Duration: Mar 14 2010Mar 19 2010

Other

Other2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010
CountryUnited States
CityDallas, TX
Period3/14/103/19/10

Fingerprint

Glossaries
Face recognition
Computer vision
MATLAB
Program processors
Availability
Data storage equipment
Processing
Graphics processing unit

Keywords

  • CUDA
  • Face recognition
  • GPU-based computing
  • Sparse representation
  • Super-resolution

ASJC Scopus subject areas

  • Signal Processing
  • Software
  • Electrical and Electronic Engineering

Cite this

Nagesh, P., Gowda, R., & Li, B. (2010). Fast GPU implementation of large scale dictionary and sparse representation based vision problems. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (pp. 1570-1573). [5495526] https://doi.org/10.1109/ICASSP.2010.5495526

Fast GPU implementation of large scale dictionary and sparse representation based vision problems. / Nagesh, Pradeep; Gowda, Rahul; Li, Baoxin.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2010. p. 1570-1573 5495526.

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

Nagesh, P, Gowda, R & Li, B 2010, Fast GPU implementation of large scale dictionary and sparse representation based vision problems. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings., 5495526, pp. 1570-1573, 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010, Dallas, TX, United States, 3/14/10. https://doi.org/10.1109/ICASSP.2010.5495526
Nagesh P, Gowda R, Li B. Fast GPU implementation of large scale dictionary and sparse representation based vision problems. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2010. p. 1570-1573. 5495526 https://doi.org/10.1109/ICASSP.2010.5495526
Nagesh, Pradeep ; Gowda, Rahul ; Li, Baoxin. / Fast GPU implementation of large scale dictionary and sparse representation based vision problems. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2010. pp. 1570-1573
@inproceedings{4c8af24f434b4a4e8837e0891241686d,
title = "Fast GPU implementation of large scale dictionary and sparse representation based vision problems",
abstract = "Recently, Computer Vision problems like Face Recognition and Super-Resolution solved using sparse representation based methods with large dictionaries have shown state-of-the-art results. However such methods are computationally prohibitive for typical CPUs, especially for a large dictionary size. We present fast implementation of these methods by exploiting the massively parallel processing capabilities of a GPU within a CUDA framework, owing to its easy off-the-shelf availability and programmer friendliness. We provide details of system level design, memory management and implementation strategies. Further, we integrate the solution to the preferred scientific computational platform - MATLAB.",
keywords = "CUDA, Face recognition, GPU-based computing, Sparse representation, Super-resolution",
author = "Pradeep Nagesh and Rahul Gowda and Baoxin Li",
year = "2010",
doi = "10.1109/ICASSP.2010.5495526",
language = "English (US)",
isbn = "9781424442966",
pages = "1570--1573",
booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

}

TY - GEN

T1 - Fast GPU implementation of large scale dictionary and sparse representation based vision problems

AU - Nagesh, Pradeep

AU - Gowda, Rahul

AU - Li, Baoxin

PY - 2010

Y1 - 2010

N2 - Recently, Computer Vision problems like Face Recognition and Super-Resolution solved using sparse representation based methods with large dictionaries have shown state-of-the-art results. However such methods are computationally prohibitive for typical CPUs, especially for a large dictionary size. We present fast implementation of these methods by exploiting the massively parallel processing capabilities of a GPU within a CUDA framework, owing to its easy off-the-shelf availability and programmer friendliness. We provide details of system level design, memory management and implementation strategies. Further, we integrate the solution to the preferred scientific computational platform - MATLAB.

AB - Recently, Computer Vision problems like Face Recognition and Super-Resolution solved using sparse representation based methods with large dictionaries have shown state-of-the-art results. However such methods are computationally prohibitive for typical CPUs, especially for a large dictionary size. We present fast implementation of these methods by exploiting the massively parallel processing capabilities of a GPU within a CUDA framework, owing to its easy off-the-shelf availability and programmer friendliness. We provide details of system level design, memory management and implementation strategies. Further, we integrate the solution to the preferred scientific computational platform - MATLAB.

KW - CUDA

KW - Face recognition

KW - GPU-based computing

KW - Sparse representation

KW - Super-resolution

UR - http://www.scopus.com/inward/record.url?scp=78049409668&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78049409668&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2010.5495526

DO - 10.1109/ICASSP.2010.5495526

M3 - Conference contribution

AN - SCOPUS:78049409668

SN - 9781424442966

SP - 1570

EP - 1573

BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

ER -