Line Correspondences from Cooperating Spatial and Temporal Grouping Processes for a Sequence of Images

Chang, Yuh-Lin; Aggarwal, J. K.

doi:10.1006/cviu.1997.0527

Cited by 16 publications

(9 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A straight line can be defined as ρ = x cos θ + y sin θ with parameter ρ, θ. So a line segment can be represented by the analytical parameters ρ, θ or position parameters, end points (x a , y a ) and (x b , y b ) or the dimensional parameter, length l [2].…”

Section: Line Segmentsmentioning

confidence: 99%

“…Now that we have a set of candidate correspondences of line segments, we must find some further constraints to determine the correct matching line segments between two images. Some criteria like similarity constraints from [2], intensity difference constraint and contour context constraint are supplied to make further matching.…”

Section: Line Matching Guided By Matched Pointsmentioning

confidence: 99%

See 1 more Smart Citation

Hierarchical Primitives Based Contour Matching

Zhang

Burkhardt²

2002

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. An algorithm for contour matching is presented in this paper. It is implemented in two steps: firstly, bottom-up, corners are matched, the matched corner points guide line segment matching, and then the matched line segments guide contour matching. Line segments are grouped with the signature function defined in [10] from the extracted contours of image pairs. Secondly, top-down, with the computed signature functions of the matched contours, the contour points are corresponded by modifying the initial matching from line segments. The novelty of our approach is that (1) features are incorporated, and the matching is implemented in two steps (bottom-up and top-down), thus dense correspondences along contours are acquired. Disadvantages such as the sparseness from only point correspondences, the inaccuracies from only line correspondences, and feature loss of correspondences along epipolar lines are avoided; (2) unlike the conventional way of matching along epipolar lines after rectification, signature functions developed in [10] are used to characterize the contours so that the matching is implemented along contours.

show abstract

Section: Line Segmentsmentioning

confidence: 99%

Section: Line Matching Guided By Matched Pointsmentioning

confidence: 99%

Hierarchical Primitives Based Contour Matching

Zhang

Burkhardt²

2002

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Therefore, the algorithm has the advantage of being performed in a parallel manner within a constant time independent of the number of features extracted from each image. On the other hand, the parallelization of irregular algorithms, such as those using feature grouping or relaxation labeling proposed, respectively, in [9] and [10], brings about several problems related to the load distribution and necessitates a mechanism that allows one to focus on the solution as soon as possible in order to eliminate useless computations. Besides the above feature, our method brings several other advantages ( Fig.…”

Section: Cam-based Ght For Segment Matchingmentioning

confidence: 99%

“…In [6], a multilevel algorithm for straight segment extraction and matching using fuzzy sets is proposed. Another alternative is to group the set of extracted segments and then match them with a model [10,11].…”

Section: Introductionmentioning

confidence: 99%

Hough Transform Algorithm for Three-Dimensional Segment Extraction and its Parallel Hardware Implementation

Meribout¹,

Nakanishi²,

Hosoya³

et al. 2000

Computer Vision and Image Understanding

View full text Add to dashboard Cite

“…Besides, many techniques have been proposed to detect general curvilinear structures such as scale space approaches, anisotropic Gauss filtering, fusion of two local line detectors followed by a global Markov random field, using differential geometric properties of images and using active contours. For the problem of matching curvilinear structures between sets of images, in the majority of the proposed methods, lines are first detected and then line properties such as orientation, position, width and center lines are used to establish correspondences [18,2,14].…”

Section: Detecting and Tracking Curvilinear Structuresmentioning

confidence: 99%