A Coarse-and-Fine Bayesian Belief Propagation for Correspondence Problems in Computer Vision

Tipwai, Preeyakorn; Madarasmi, Suthep

doi:10.1007/978-3-540-76631-5_65

Cited by 3 publications

(2 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is shown that this transformation at the zero temperature limit is identically the same as RGT [42]. There are some other multiscale approaches such as [43,44] in the literature. The technique employed here applies multiresolution ideas in a principled way, based on supercoupling transform.…”

Section: Multi-scale Relaxationmentioning

confidence: 85%

Pose-invariant face recognition by matching on multi-resolution MRFs linked by supercoupling transform

Arashloo

Kittler

Christmas

2011

Computer Vision and Image Understanding

View full text Add to dashboard Cite

Section: Multi-scale Relaxationmentioning

confidence: 85%

Pose-invariant face recognition by matching on multi-resolution MRFs linked by supercoupling transform

Arashloo

Kittler

Christmas

2011

Computer Vision and Image Understanding

View full text Add to dashboard Cite

“…where ρ n is the maximum distance at level n. (15) where m pq is m below1 and m below2 when q are two of (2x, 2y), (2x+1, 2y), (2x, 2y+1) or (2x+1, 2y+1), which are edge pixels on level n+1.…”

Section: Bp I: Context Searchmentioning

confidence: 99%

A dual belief propagation method for shape recognition

Tipwai

Madarasmi

2009

2009 IEEE Symposium on Computational Intelligence for Image Processing

Self Cite

View full text Add to dashboard Cite

We present a shape recognition framework which includes two steps: shape searching and shape matching by deformation. First, the user can draw a contour shape descriptor as a search template. The first Bayesian belief propagation (BP I) algorithm is used to find possible targets allowing for translation, scale, and rotation transformations to all contours in a cluttered image. The contour segments with common transformation values are grouped and hypothesized as belonging to the contour in the search template. The search template is then transformed for each possible transformation value. A second belief propagation (BP II) is applied to perform a deformable contour matching. The matching score or cost function determines whether there is an actual match. The algorithm overcomes the weaknesses of the other approaches since it does not require any pre-processing to detect feature points, it can match targets at any position, scale, or rotation transformations, and it does not use any accumulation space that my have peak clustering problems such as in the Hough Transform.P. Tipwai is with the

show abstract

Object recognition by belief propagation

Sang²,

Liu³

et al. 2008

Opt. Eng

View full text Add to dashboard Cite

A Coarse-and-Fine Bayesian Belief Propagation for Correspondence Problems in Computer Vision

Cited by 3 publications

References 10 publications

Pose-invariant face recognition by matching on multi-resolution MRFs linked by supercoupling transform

Pose-invariant face recognition by matching on multi-resolution MRFs linked by supercoupling transform

A dual belief propagation method for shape recognition

Object recognition by belief propagation

Contact Info

Product

Resources

About