2001
DOI: 10.1109/76.905989
|View full text |Cite
|
Sign up to set email alerts
|

Fast full-search block matching

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
43
0
1

Year Published

2002
2002
2018
2018

Publication Types

Select...
5
3
1

Relationship

0
9

Authors

Journals

citations
Cited by 87 publications
(44 citation statements)
references
References 10 publications
0
43
0
1
Order By: Relevance
“…The motion measurement technology based on digital image has developed more than 30 years, there were many estimation algorithms during these years, such as matching method [1], grey projection method [2], phase correlation method [3] and so on. SIFT (Scale-invariant feature transform) method is the most widely used and its effect is the most stable motion estimation algorithm based on feature.…”
Section: Methods Based On Imagementioning
confidence: 99%
“…The motion measurement technology based on digital image has developed more than 30 years, there were many estimation algorithms during these years, such as matching method [1], grey projection method [2], phase correlation method [3] and so on. SIFT (Scale-invariant feature transform) method is the most widely used and its effect is the most stable motion estimation algorithm based on feature.…”
Section: Methods Based On Imagementioning
confidence: 99%
“…Additionally, to save compu− tations, the prediction of the starting level is employed, which utilizes information about the rejection level of the preceding block. Other modifications of SEA are: extended SEA (ESEA) [68], extended MSEA (EMSEA) [69] and adaptive MSEA (AdaMSEA) [70].…”
Section: Fast Full-search Algorithmsmentioning
confidence: 99%
“…For example, reference [11] is the extension of [5]. Reference [12] is proposed based on [9], and reference [13] combined [10] with multilevel successive elimination algorithm [14], [15]. Reference [16] is the extension of [6].…”
Section: Impact Of Supporting Vbsme In Different Hardware Architementioning
confidence: 99%
“…A moving window is also adopted to save the computation complexity and reduce on-chip memory bandwidth. The search range is partitioned into four regions, i.e., [ 64,15], [ 48,31], [ 32,47], and [ 16,63] in the horizontal direction. According to an MV predictor, one of four regions is selected and the searching candidates in this region are computed.…”
Section: M-parallel Sad Treementioning
confidence: 99%