Graph Matching via Sequential Monte Carlo

Suh, Young-Ger; Cho, Minsu; Lee, Kyoung Mu

doi:10.1007/978-3-642-33712-3_45

Cited by 31 publications

(18 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2(b). These are "soft" constraints (e.g., as opposed to ones in [20,24]), the consistent elements of x being driven toward a solution satisfying these constraints (e.g., as similar to ones in [6,10,14]). MPM itself does not prevent one-to-many nor many-to-one matches if they are well supported by maxpooled neighboring matches.…”

Section: Algorithm 1: Max-pooling Matching (Mpm)mentioning

confidence: 99%

See 1 more Smart Citation

Finding Matches in a Haystack: A Max-Pooling Strategy for Graph Matching in the Presence of Outliers

Cho

Sun

Duchenne³

et al. 2014

2014 IEEE Conference on Computer Vision and Pattern Recognition

Self Cite

125

View full text Add to dashboard Cite

A major challenge in real-world feature matching problems is to tolerate the numerous outliers arising in typical visual tasks. Variations in object appearance, shape, and structure within the same object class make it harder to distinguish inliers from outliers due to clutters. In this paper, we propose a max-pooling approach to graph matching, which is not only resilient to deformations but also remarkably tolerant to outliers. The proposed algorithm evaluates each candidate match using its most promising neighbors, and gradually propagates the corresponding scores to update the neighbors. As final output, it assigns a reliable score to each match together with its supporting neighbors, thus providing contextual information for further verification. We demonstrate the robustness and utility of our method with synthetic and real image experiments.

show abstract

Section: Algorithm 1: Max-pooling Matching (Mpm)mentioning

confidence: 99%

“…More precisely, it is the quadratic assignment problem, which is known to be NP-hard. Due to its generality and flexibility, this formulation and its extensions has been favored in recent graph matching research [6,7,10,18,20,24,31,32]. …”

Section: Standard Formulationmentioning

confidence: 99%

Finding Matches in a Haystack: A Max-Pooling Strategy for Graph Matching in the Presence of Outliers

Cho

Sun

Duchenne³

et al. 2014

2014 IEEE Conference on Computer Vision and Pattern Recognition

Self Cite

125

View full text Add to dashboard Cite

show abstract

“…• pairwise visual object matching which matches feature point correspondences between a pair of objects [15], [11], [4]; and…”

Section: Related Workmentioning

confidence: 99%

Planogram Compliance Checking Using Recurring Patterns

Liu

Tian

2015

2015 IEEE International Symposium on Multimedia (ISM)

View full text Add to dashboard Cite

This paper proposes a novel automated planogram compliance checking method for retail chains without requiring product template images for modeling or training. Product layout information is extracted from one single input image by means of unsupervised recurring pattern detection and matched via graph matching with the expected product layout specified by a planogram. To improve the efficiency, a divide-conquer strategy is employed. Specifically, the input image is divided into several regions based on the planogram. Recurring patterns are detected in each region respectively and then merged together to estimate product layout information. Experimental results on real data from a supermarket chain have verified the effectiveness and efficiency of the proposed method.

show abstract

“…Two classic approaches for recurring pattern detection are: (A) pairwise visual-word-matching which matches pairs of visual words across all objects [7]; and (B) pairwise object-matching which matches feature point correspondences between a pair of objects [12,5,4]. Both of these methods are limited in that (1) Pairwise matching, though relatively simple, does not fully utilize all available information for optimal matching.…”

Section: Introductionmentioning

confidence: 99%

GRASP Recurring Patterns from a Single View

Liu

2013

2013 IEEE Conference on Computer Vision and Pattern Recognition

View full text Add to dashboard Cite

We propose a novel unsupervised method for discovering recurring patterns from a single view. A key contribution of our approach is the formulation and validation of a joint assignment optimization problem where multiple visual words and object instances of a potential recurring pattern are considered simultaneously. The optimization is achieved by a greedy randomized adaptive search procedure (GRASP) with moves specifically designed for fast convergence. We have quantified systematically the performance of our approach under stressed conditions of the input (missing features, geometric distortions). We demonstrate that our proposed algorithm outperforms state of the art methods for recurring pattern discovery on a diverse set of 400+ real world and synthesized test images.

show abstract

Graph Matching via Sequential Monte Carlo

Cited by 31 publications

References 25 publications

Finding Matches in a Haystack: A Max-Pooling Strategy for Graph Matching in the Presence of Outliers

Finding Matches in a Haystack: A Max-Pooling Strategy for Graph Matching in the Presence of Outliers

Planogram Compliance Checking Using Recurring Patterns

GRASP Recurring Patterns from a Single View

Contact Info

Product

Resources

About