Incremental Multi-graph Matching via Diversity and Randomness Based Graph Clustering

Yu, Tianshu; Yan, Junchi; Liu, Wei; Li, Baoxin

doi:10.1007/978-3-030-01261-8_9

Cited by 15 publications

(8 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The other line of methods impose cycle-consistency during the iterative finding of pairwise matchings, and usually can achieve better results [36,33,35,34]. Another relevant setting is solving multiple graph matching in an online fashion [38].…”

Section: Learning-free Graph Matching Methodsmentioning

confidence: 99%

Neural Graph Matching Network: Learning Lawler's Quadratic Assignment Problem with Extension to Hypergraph and Multiple-graph Matching

Wang

Yan

Yang

2019

Preprint

Self Cite

View full text Add to dashboard Cite

Graph matching involves combinatorial optimization based on edge-to-edge affinity matrix, which can be generally formulated as Lawlers Quadratic Assignment Problem (QAP). This paper presents a QAP network directly learning with the affinity matrix (equivalently the association graph) whereby the matching problem is translated into a vertex classification task. The association graph is learned by an embedding network for vertex classification, followed by Sinkhorn normalization and a cross-entropy loss for endto-end learning. We further improve the embedding model on association graph by introducing Sinkhorn based constraint, and dummy nodes to deal with outliers. To our best knowledge, this is the first network to directly learn with the general Lawlers QAP. In contrast, state-of-the-art deep matching methods [39,32,42] focus on the learning of node and edge features in two graphs respectively. We also show how to extend our network to hypergraph matching, and matching of multiple graphs. Experimental results on both synthetic graphs and real-world images show our method outperforms. For pure QAP tasks on synthetic data and QAPLIB, our method can surpass spectral matching [19] and RRWM [9], especially on challenging problems.

show abstract

Section: Learning-free Graph Matching Methodsmentioning

confidence: 99%

Neural Graph Matching Network: Learning Lawler's Quadratic Assignment Problem with Extension to Hypergraph and Multiple-graph Matching

Wang

Yan

Yang

2019

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Note in this work, it is assumed that all the graphs still fall into one single cluster, and the partition is enforced in brute-force. In (Yu et al 2018), incremental matching of graph sequence is addressed, as inspired by (Hu, Thibert, and Guibas 2018).…”

Section: Related Workmentioning

confidence: 99%

“…First we follow the tradition in previous multi-graph matching literature (Yan et al 2015a;2016a;Yu et al 2018) which define the term supergraph as follows.…”

Section: Multi-graph Matching and Clustering Preliminariesmentioning

confidence: 99%

See 1 more Smart Citation

Multiple Graph Matching and Clustering via Decayed Pairwise Matching Composition

Wang

Jiang

Yan

2020

AAAI

Self Cite

View full text Add to dashboard Cite

Jointly matching of multiple graphs is challenging and recently has been an active topic in machine learning and computer vision. State-of-the-art methods have been devised, however, to our best knowledge there is no effective mechanism that can explicitly deal with the matching of a mixture of graphs belonging to multiple clusters, e.g., a collection of bikes and bottles. Seeing its practical importance, we propose a novel approach for multiple graph matching and clustering. Firstly, for the traditional multi-graph matching setting, we devise a composition scheme based on a tree structure, which can be seen as in the between of two strong multi-graph matching solvers, i.e., MatchOpt (Yan et al. 2015a) and CAO (Yan et al. 2016a). In particular, it can be more robust than MatchOpt against a set of diverse graphs and more efficient than CAO. Then we further extend the algorithm to the multiple graph matching and clustering setting, by adopting a decaying technique along the composition path, to discount the meaningless matching between graphs in different clusters. Experimental results show the proposed methods achieve excellent trade-off on the traditional multi-graph matching case, and outperform in both matching and clustering accuracy, as well as time efficiency.

show abstract

“…As an important task in unsupervised learning [43,8,22,24] and vision communities [48], clustering [14] has been widely used in image segmentation [37], image categorization [45,47], and digital media analysis [1]. The goal of clustering is to find a partition in order to keep similar data points in the same cluster while dissimilar ones in different clusters.…”

Section: Introductionmentioning

confidence: 99%

Deep Spectral Clustering Using Dual Autoencoder Network

Yang

Deng

Zheng

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

Self Cite

229

View full text Add to dashboard Cite

The clustering methods have recently absorbed evenincreasing attention in learning and vision. Deep clustering combines embedding and clustering together to obtain optimal embedding subspace for clustering, which can be more effective compared with conventional clustering methods. In this paper, we propose a joint learning framework for discriminative embedding and spectral clustering. We first devise a dual autoencoder network, which enforces the reconstruction constraint for the latent representations and their noisy versions, to embed the inputs into a latent space for clustering. As such the learned latent representations can be more robust to noise. Then the mutual information estimation is utilized to provide more discriminative information from the inputs. Furthermore, a deep spectral clustering method is applied to embed the latent representations into the eigenspace and subsequently clusters them, which can fully exploit the relationship between inputs to achieve optimal clustering results. Experimental results on benchmark datasets show that our method can significantly outperform state-of-the-art clustering approaches.

show abstract

Incremental Multi-graph Matching via Diversity and Randomness Based Graph Clustering

Cited by 15 publications

References 35 publications

Neural Graph Matching Network: Learning Lawler's Quadratic Assignment Problem with Extension to Hypergraph and Multiple-graph Matching

Neural Graph Matching Network: Learning Lawler's Quadratic Assignment Problem with Extension to Hypergraph and Multiple-graph Matching

Multiple Graph Matching and Clustering via Decayed Pairwise Matching Composition

Deep Spectral Clustering Using Dual Autoencoder Network

Contact Info

Product

Resources

About