Learning Graph Matching

By Tibério S. Caetano, Li Cheng, Quoc V. Le, Alex Smola. 2007

465 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Graph Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Graph Learning

Citations in the corpus

Block-Coordinate Frank-Wolfe Optimization for Structural SVMs (2012, methodology): “, 2003), graph matchings (Caetano et al., 2009), and associative Markov networks (Taskar, 2004).”
Learning Attitudes and Attributes from Multi-aspect Reviews (2012, methodology): “The use of bipartite graph cover objectives in structured learning is addressed in [6], where an objective similar to that of (eq.”
Kernel Methods in Computer Vision (2009, background): “[Caetano et al., 2007] propose a structure learning formulation based on the S-SVM for this task, where the two sets of landmark points act as input while the correspondence assignment between the sets is the output.”
SuperGlue: Learning Feature Matching With Graph Neural Networks (2019, methodology): “[8] learn the cost of the optimization for a simpler linear assignment, but only use a shallow model, while our SuperGlue learns a flexible cost using a deep neural network.”
Learning Latent Permutations with Gumbel-Sinkhorn Networks (2018, methodology): “Prior work has developed learning algorithms for supervised learning where the training data includes annotated matchings (Caetano et al., 2009; Petterson et al., 2009; Tang et al., 2016).”
A review of multimodal image matching: Methods and applications (2021, background): “Affinity matrix plays a key role in the GM problem, which is why most deep learning techniques tend to obtain a better representation for it in a supervised [275] or unsupervised [276] form.”
Attributed Graph Mining and Matching: An Attempt to Define and Extract Soft Attributed Patterns (2014, methodology): “Views of graph matching: Given a graph template and a number of ARGs, conventional algorithms for learning graph matching [4, 16, 17, 24] mainly train matching parameters, and Cho et al.”
Discriminative prototype selection methods for graph embedding (2013, background): “This section briefly conveys the key terminology and some concepts used in this paper.”
Category Modeling from Just a Single Labeling: Use Depth Information to Guide the Learning of 2D Models (2013, background): “Conventional algorithms for learning graph matching [2, 3, 4] have focused on training the weights of different graphical attributes, given a template graph (the category model) and multiple target graphs.”
Kernelized Sorting (2008, unknown): “Index Terms—Sorting, matching, kernels, object alignment, Hilbert-Schmidt Independence Criterion.”
Solving the multi-way matching problem by permutation synchronization (2013, methodology): “Instead of learning parameters [24, 26], we utilize supervision directly to provide the correct matches on a small subset of randomly picked image pairs (e.”
A tensor-based algorithm for high-order graph matching (2009, unknown): “Moreover, it is natural to follow the approach of [6] to learn potentials automatically from labelled or partially labelled data.”
Feature Correspondence Via Graph Matching: Models and Global Optimization (2008, methodology): “Recent work [5] has suggested that for many corr espondence problems, learned appearance-based models perform similarly or bett er than state-of-the-art structural models exploiting information about spatial arrange ment of features.”
Image Matching from Handcrafted to Deep Features: A Survey (2020, methodology): “To this end, unsupervised (Leordeanu et al. 2012) and supervised (Caetano et al. 2009) methods are devised to learn K. Zanfir and Sminchisescu (2018) recently addressed this issue with an end-to-end deep learning scheme.”
Factorized Graph Matching (2016, background): “In addition, we are exploring the extension of FGM to other higher-order graph matching problems [42], [46] as well as learning parameters for graph matching [44], [45].”