Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Skip to main content
We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate
arXiv logo
Cornell University Logo

Computer Science > Computer Vision and Pattern Recognition

arXiv:0806.0689 (cs)
[Submitted on 4 Jun 2008]

Title:Directional Cross Diamond Search Algorithm for Fast Block Motion Estimation

Authors:Hongjun Jia, Li Zhang
View PDF
Abstract: In block-matching motion estimation (BMME), the search patterns have a significant impact on the algorithm's performance, both the search speed and the search quality. The search pattern should be designed to fit the motion vector probability (MVP) distribution characteristics of the real-world sequences. In this paper, we build a directional model of MVP distribution to describe the directional-center-biased characteristic of the MVP distribution and the directional characteristics of the conditional MVP distribution more exactly based on the detailed statistical data of motion vectors of eighteen popular sequences. Three directional search patterns are firstly designed by utilizing the directional characteristics and they are the smallest search patterns among the popular ones. A new algorithm is proposed using the horizontal cross search pattern as the initial step and the horizontal/vertical diamond search pattern as the subsequent step for the fast BMME, which is called the directional cross diamond search (DCDS) algorithm. The DCDS algorithm can obtain the motion vector with fewer search points than CDS, DS or HEXBS while maintaining the similar or even better search quality. The gain on speedup of DCDS over CDS or DS can be up to 54.9%. The simulation results show that DCDS is efficient, effective and robust, and it can always give the faster search speed on different sequences than other fast block-matching algorithm in common use.
Comments: 23 pages, 9 figures, technical report, related paper under submission
Subjects: Computer Vision and Pattern Recognition (cs.CV)
ACM classes: I.4.2
Cite as: arXiv:0806.0689 [cs.CV]
  (or arXiv:0806.0689v1 [cs.CV] for this version)
  https://doi.org/10.48550/arXiv.0806.0689

Submission history

From: Hongjun Jia [view email]
[v1] Wed, 4 Jun 2008 05:05:19 UTC (294 KB)
Full-text links:

Access Paper:

  • View PDF
  • Other Formats
view license
Current browse context:
cs.CV
< prev   |   next >
new | recent | 2008-06
Change to browse by:
cs

References & Citations

DBLP - CS Bibliography

listing | bibtex
Hongjun Jia
Li Zhang
export BibTeX citation

Bookmark

BibSonomy logo Reddit logo

Bibliographic and Citation Tools

Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)

Code, Data and Media Associated with this Article

alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)

Demos

Replicate (What is Replicate?)
Hugging Face Spaces (What is Spaces?)
TXYZ.AI (What is TXYZ.AI?)

Recommenders and Search Tools

Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)