Implementation of a fast image coding and retrieval system using a GPU

Author

Sattigeri, Prasanna ; Thiagarajan, Jayaraman J. ; Ramamurthy, Karthikeyan N. ; Spanias, Andreas

Author_Institution

SenSIP Center & Ind. Consortium, Arizona State Univ., Tempe, AZ, USA

fYear

2012

fDate

12-14 Jan. 2012

Firstpage

5

Lastpage

8

Abstract

Sparse coding of image patches is a compact but computationally expensive method of representing images. As part of our SenSIP consortium industry projects, we implement the Orthogonal Matching Pursuit algorithm using a single CUDA kernel on a GPU and sparse codes for image patches are obtained in parallel. Image-based “exact search” and “visually similar search” using the image patch sparse codes are performed. Results demonstrate large speed-up over CPU implementations and good retrieval performance is also achieved.

Keywords

graphics processing units; image coding; image representation; image retrieval; CPU implementations; GPU; SenSIP consortium industry projects; fast image coding; image patch sparse codes; image patches; image representation; image-based exact search; image-based visually similar search; orthogonal matching pursuit algorithm; retrieval system; single CUDA kernel; sparse coding; Dictionaries; Graphics processing unit; Image coding; Image retrieval; Kernel; Matching pursuit algorithms; Vectors; GPU implementation; Sparse coding; image retrieval; orthogonal matching pursuit;

fLanguage

English

Publisher

ieee

Conference_Titel

Emerging Signal Processing Applications (ESPA), 2012 IEEE International Conference on

Conference_Location

Las Vegas, NV

Print_ISBN

978-1-4673-0899-1

Type

conf

DOI

10.1109/ESPA.2012.6152431

Filename

6152431