2006
DOI: 10.1007/978-3-540-69429-8_2
|View full text |Cite
|
Sign up to set email alerts
|

Exploiting Video Stream Similarity for Energy-Efficient Decoding

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
5
0

Year Published

2007
2007
2020
2020

Publication Types

Select...
5
1

Relationship

1
5

Authors

Journals

citations
Cited by 7 publications
(5 citation statements)
references
References 15 publications
0
5
0
Order By: Relevance
“…Scenarioaware resource prediction can also be used to predict the energy consumed by such energy-efficient video decoding solutions. Figure 6 shows the error when predicting the amount of energy consumed for decoding the given video stream using the DVFS-driven video decoding method presented in [10]. The average prediction error is 1.4%; the maximum prediction error is never larger than 4%.…”
Section: Predicting Energy Consumptionmentioning
confidence: 99%
“…Scenarioaware resource prediction can also be used to predict the energy consumed by such energy-efficient video decoding solutions. Figure 6 shows the error when predicting the amount of energy consumed for decoding the given video stream using the DVFS-driven video decoding method presented in [10]. The average prediction error is 1.4%; the maximum prediction error is never larger than 4%.…”
Section: Predicting Energy Consumptionmentioning
confidence: 99%
“…Several video decoding adaption frameworks that combine with power-aware software and hardware modules to optimize battery life have been proposed. Dynamic frequency/voltage scaling (DVS) [1,2] and workload reshaping (WR) [3][4][5] are the most attented methodologies for both industry and academic.…”
Section: Introductionmentioning
confidence: 99%
“…Since the implementation cost of an exact LRU policy can become prohibitive, an approximation of LRU is usually implemented. Other replacement policies include first in first out (FIFO) [Hayes 1988] and Belady's optimal (OPT) replacement policy [Belady 1966]. The latter presumes a full knowledge of all the future accesses and then always discards the data which will not be needed for the longest time.…”
Section: Generalization Of Resultsmentioning
confidence: 99%
“…In [Bernabe 2005], cache misses are reduced by tile processing but without taking into consideration the impact of the input variation. Other content-driven mapping techniques also reduce the energy consumption by minimizing memory misses [Hamers 2007] and specifying optimal execution orders under profiled sets of operating conditions or scenarios [Gomez 2002]. However, these examples exploit more coarse-grain features of the content variation and they are not efficient for high-complexity WT-based applications.…”
Section: Memory Hierarchy Mappingmentioning
confidence: 99%
See 1 more Smart Citation