Paper Info
Title
Energy scalable approximate DCT architecture trading quality via boundary error-resiliency
Abstract
High energy-efficiency is an imperative requirement for today's portable multimedia devices. Most image/video processing devices employ energy-devouring Discrete Cosine Transform (DCT) unit that limits their performance. We propose a novel low-complexity DCT architecture that exploits non-significance-driven complexity reduction techniques namely neighbor pixel similarity (NPS) and constant multiplication factor (CMF). Further, quality-energy tradeoff is achieved via error-resiliency at image-boundaries. We also present a novel quality metric PMSE (Perceivable Mean Square Error) that quantifies perceivable error. Our architecture consumes 49.2% less energy at the expense of 1.43dB less PSNR and requires only 45.4% area over well-known DCT architectures.
Year
DOI
Venue
2014
10.1109/SOCC.2014.6948945
System-on-Chip Conference
Keywords
Field
DocType
discrete cosine transforms,energy conservation,mean square error methods,multimedia communication,video signal processing,CMF,DCT architecture trading quality,DCT unit,NPS,PMSE,PSNR,boundary error-resiliency,constant multiplication factor,discrete cosine transform unit,energy efficiency,image boundaries,image processing devices,neighbor pixel similarity,nonsignificance-driven complexity reduction techniques,perceivable mean square error,portable multimedia devices,quality-energy tradeoff,video processing devices,Discrete Cosine Transform (DCT),architecture,error-resiliency,quality-energy tradeoff
Architecture,Video processing,Computer science,Discrete cosine transform,Mean squared error,Real-time computing,Reduction (complexity),Multiplication,Pixel,Computer engineering,Scalability
Conference
Citations 
PageRank 
References 
1
0.37
7
Authors
3
Name
Order
Citations
PageRank
Bharat Garg1329.88
Nitesh K. Bharadwaj211.38
Sharma, G.K.371.87