2011
DOI: 10.1016/j.image.2011.01.001
|View full text |Cite
|
Sign up to set email alerts
|

A high-throughput ASIC processor for 8×8 transform coding in H.264/AVC

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
2

Citation Types

0
4
0

Year Published

2011
2011
2017
2017

Publication Types

Select...
2
2
1

Relationship

0
5

Authors

Journals

citations
Cited by 5 publications
(5 citation statements)
references
References 22 publications
0
4
0
Order By: Relevance
“…However, the quantization and rescaling are computed using only one multiplier each and they are performed at the pace demanded by the entropy coder. In a previous work (Michell et al, 2011), we described a parallel architecture capable of processing 8×8 blocks without interruption with a bit-depth fixed to 8 bit. The latency of 38 clock cycles is achieved by implementing in a pipeline scheme each module used in the transform coding.…”
Section: Asic Implementation and Comparisonsmentioning
confidence: 99%
“…However, the quantization and rescaling are computed using only one multiplier each and they are performed at the pace demanded by the entropy coder. In a previous work (Michell et al, 2011), we described a parallel architecture capable of processing 8×8 blocks without interruption with a bit-depth fixed to 8 bit. The latency of 38 clock cycles is achieved by implementing in a pipeline scheme each module used in the transform coding.…”
Section: Asic Implementation and Comparisonsmentioning
confidence: 99%
“…In a previous work (Michell et al, 2011), we described a parallel architecture capable of processing 8×8 blocks without interruption with a bit-depth fixed to 8 bit. The latency of 38 clock cycles is achieved by implementing in a pipeline scheme each module used in the transform coding.…”
Section: Asic Implementation and Comparisonsmentioning
confidence: 99%
“…It achieves substantially better coding efficiency compared to its predecessors by means of a larger number of coding tools, such as new prediction modes, larger transform sizes and new picture partitioning, which require a much higher computational effort with respect to the previous standards [3][4][5][6]. As with previous standards, custom hardware implementations can accelerate some operations [7][8][9][10]. Such hardware is typically integrated in many devices, including cameras and mobile phones.…”
Section: Introductionmentioning
confidence: 99%