2005
DOI: 10.1117/12.633389
|View full text |Cite
|
Sign up to set email alerts
|

Statistics based fast intra mode detection

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3

Citation Types

0
3
0
1

Year Published

2006
2006
2013
2013

Publication Types

Select...
4
3

Relationship

0
7

Authors

Journals

citations
Cited by 7 publications
(4 citation statements)
references
References 0 publications
0
3
0
1
Order By: Relevance
“…Fast intra mode decision algorithms using edge detection histogram and local edge detection were proposed in [3], [4]. Also, there exist fast algorithms using statistics-based methods such as [5]. Most of these algorithms speed up the encoding time, but result in slight PSNR degradation, because those cannot always decide the best mode.…”
Section: Introductionmentioning
confidence: 99%
“…Fast intra mode decision algorithms using edge detection histogram and local edge detection were proposed in [3], [4]. Also, there exist fast algorithms using statistics-based methods such as [5]. Most of these algorithms speed up the encoding time, but result in slight PSNR degradation, because those cannot always decide the best mode.…”
Section: Introductionmentioning
confidence: 99%
“…To reduce the number of candidates, edge detection techniques are employed to decide the edge direction and the number of modes to investigate is reduced. Other methods to limit the number of candidates are based on filters [24], directional mask [25], intensity gradient [26], and statistical properties [27]. Limiting the number of candidates reduce the encoding time, however, the preprocessing to detect the edges and classify them requires the additional encoding time and reduces the coding efficiency.…”
Section: Introductionmentioning
confidence: 99%
“…An edge detection technique was employed in [15] to determine the edge direction, and, based on the direction, the number of candidate modes for the best mode was limited. Other approaches for limiting the number of candidates were based on filters [20], [21], a directional mask [22], the intensity gradient [23], and statistical properties [24]. Although these approaches reduce the encoding complexity, such preprocessing as detecting the edges or classifying their directional patterns requires an additional computational burden on the encoder.…”
Section: Introductionmentioning
confidence: 99%