Sciweavers

7 search results - page 1 / 2
» Avoiding Conversion and Rearrangement Overhead in SIMD Archi...
Sort
View
IJPP
2006
61views more  IJPP 2006»
13 years 7 months ago
Avoiding Conversion and Rearrangement Overhead in SIMD Architectures
Asadollah Shahbahrami, Ben H. H. Juurlink, Demid B...
DSD
2009
IEEE
148views Hardware» more  DSD 2009»
14 years 2 months ago
SIMD Architectural Enhancements to Improve the Performance of the 2D Discrete Wavelet Transform
—The 2D Discrete Wavelet Transform (DWT) is a time-consuming kernel in many multimedia applications such as JPEG2000 and MPEG-4. The 2D DWT consists of horizontal filtering alon...
Asadollah Shahbahrami, Ben H. H. Juurlink
ISM
2006
IEEE
130views Multimedia» more  ISM 2006»
14 years 1 months ago
Accelerating Color Space Conversion Using Extended Subwords and the Matrix Register File
Color space conversion is an important kernel in multimedia codecs such as JPEG and MPEG. When implemented using SIMD instructions, however, the performance improvement is often l...
Asadollah Shahbahrami, Ben H. H. Juurlink, Stamati...
CASES
2006
ACM
14 years 1 months ago
Limitations of special-purpose instructions for similarity measurements in media SIMD extensions
Microprocessor vendors have provided special-purpose instructions such as psadbw and pdist to accelerate the sumof-absolute differences (SAD) similarity measurement. The usefulne...
Asadollah Shahbahrami, Ben H. H. Juurlink, Stamati...
CF
2005
ACM
13 years 9 months ago
Matrix register file and extended subwords: two techniques for embedded media processors
In this paper we employ two techniques suitable for embedded media processors. The first technique, extended subwords, uses four extra bits for every byte in a media register. Th...
Asadollah Shahbahrami, Ben H. H. Juurlink, Stamati...