HiPPI: Higher-Order Projected Power Iterations for Scalable Multi-Matching

@article{Bernard2019HiPPIHP,
  title={HiPPI: Higher-Order Projected Power Iterations for Scalable Multi-Matching},
  author={Florian Bernard and Johan Thunberg and Paul Swoboda and Christian Theobalt},
  journal={2019 IEEE/CVF International Conference on Computer Vision (ICCV)},
  year={2019},
  pages={10283-10292}
}
The matching of multiple objects (e.g. shapes or images) is a fundamental problem in vision and graphics. In order to robustly handle ambiguities, noise and repetitive patterns in challenging real-world settings, it is essential to take geometric consistency between points into account. Computationally, the multi-matching problem is difficult. It can be phrased as simultaneously solving multiple (NP-hard) quadratic assignment problems (QAPs) that are coupled via cycle-consistency constraints… 

Figures and Tables from this paper

Isometric Multi-Shape Matching
TLDR
A novel optimisation formulation for isometric multi-shape matching that obtains multi-matchings that are by construction provably cycle-consistent and provides a convergence and complexity analysis.
Scalable Cluster-Consistency Statistics for Robust Multi-Object Matching
TLDR
New statistics for robustly filtering corrupted key point matches in the structure from motion pipeline are developed based on consistency constraints that arise within the clustered structure of the graph of keypoint matches.
Image Matching from Handcrafted to Deep Features: A Survey
TLDR
This survey introduces feature detection, description, and matching techniques from handcrafted methods to trainable ones and provides an analysis of the development of these methods in theory and practice, and briefly introduces several typical image matching-based applications.
Joint Deep Multi-Graph Matching and 3D Geometry Learning from Inhomogeneous 2D Image Collections
TLDR
This work proposes a trainable framework that takes advantage of graph neu- ral networks for learning a deformable 3D geometry model from inhomogeneous image collections, i.e., a set of images that depict different instances of objects from the same category.
MINA: Convex Mixed-Integer Programming for Non-Rigid Shape Alignment
TLDR
This work proposes a novel shape deformation model based on an efficient low-dimensional discrete model so that finding a globally optimal solution is tractable in (most) practical cases and combines several favourable properties.
Multi-frame Motion Segmentation by Combining Two-Frame Results
TLDR
This paper introduces a two-stage approach to motion segmentation on sparse and unstructured datasets involving rigid motions, motivated by multibody structure from motion, and assumes only two-frame correspondences as input without prior knowledge about trajectories.
Graduated Assignment for Joint Multi-Graph Matching and Clustering with Application to Unsupervised Graph Matching Network Learning
TLDR
Experimental results on real-world benchmarks show the graduated assignment procedure for soft matching and clustering over iterations outperforms learning-free algorithms and performs comparatively against two-graph based supervised graph matching approaches.
Variational Autoencoders for Localized Mesh Deformation Component Analysis
TLDR
A mesh-based variational autoencoder architecture that is able to cope with meshes with irregular connectivity and nonlinear deformations, assuming that the analyzed dataset contains meshes with the same vertex connectivity, which is common for deformation analysis is proposed.
Sparse Quadratic Optimisation over the Stiefel Manifold with Application to Permutation Synchronisation
TLDR
This work proposes a simple yet effective sparsity-promoting modification of the Orthogonal Iteration algorithm for finding the dominant eigenspace of a matrix that can guarantee that this method finds a Stiefel matrix that is globally optimal with respect to the quadratic objective function, while in addition being sparse.
...
...

References

SHOWING 1-10 OF 71 REFERENCES
Multi-image Matching via Fast Alternating Minimization
TLDR
The applicability of the proposed method to match the images of different object instances and as a result the potential to reconstruct category-specific object models from those images is demonstrated.
Near-Optimal Joint Object Matching via Convex Relaxation
TLDR
An algorithm to jointly match multiple objects that exhibit only partial similarities, given a few pairwise matches that are densely corrupted, is developed called MatchLift, following a spectral method that pre-estimates the total number of distinct elements to be matched.
A flexible tensor block coordinate ascent scheme for hypergraph matching
TLDR
This paper proposes a general framework for tensor block coordinate ascent methods for hypergraph matching and proposes two algorithms which both come along with the guarantee of monotonic ascent in the matching score on the set of discrete assignment matrices.
Fast Multi-image Matching via Density-Based Clustering
TLDR
This paper proposes QuickMatch, a novel algorithm that identifies multi-image matches from a density function in feature space, and uses the density estimate to order the points in a tree, and then extracts the matches by breaking this tree using feature distances and measures of distinctiveness.
Factorized Graph Matching
TLDR
Factorized graph matching (FGM) is proposed, which factorizes the large pairwise affinity matrix into smaller matrices that encode the local structure of each graph and the Pairwise affinity between edges.
Graduated Consistency-Regularized Optimization for Multi-graph Matching
TLDR
This paper proposes a novel multi-graph matching algorithm to incorporate the two aspects by iteratively approximating the global-optimal affinity score, meanwhile gradually infusing the consistency as a regularizer, which improves the performance of the initial solutions obtained by existing pairwise graph matching solvers.
A Dual Decomposition Approach to Feature Correspondence
TLDR
A novel graph matching optimization technique, which is referred to as dual decomposition (DD), is described, and it is demonstrated on a variety of examples that this method outperforms existing graph matching algorithms.
A Global Solution to Sparse Correspondence Problems
TLDR
A new methodology for reliably solving the correspondence problem between sparse sets of points of two or more images is proposed, which performs correspondence and outlier rejection in a single step and achieves global optimality with feasible computation.
Practical and Efficient Multi-view Matching
TLDR
A novel solution to the multi-view matching problem that, given a set of noisy pairwise correspondences, jointly updates them so as to maximize their consistency is proposed, based on a spectral decomposition.
The Projected Power Method: An Efficient Algorithm for Joint Alignment from Pairwise Differences
TLDR
This paper proposes a low‐complexity and model‐free nonconvex procedure, which operates in a lifted space by representing distinct label values in orthogonal directions and attempts to optimize quadratic functions over hypercubes.
...
...