Abstract:Accurately matching local features between a pair of images is a challenging computer vision task. Previous studies typically use attention based graph neural networks (GNNs) with fully-connected graphs over keypoints within/across images for visual and geometric information reasoning. However, in the context of feature matching, considerable keypoints are non-repeatable due to occlusion and failure of the detector, and thus irrelevant for message passing. The connectivity with non-repeatable keypoints not only introduces redundancy, resulting in limited efficiency, but also interferes with the representation aggregation process, leading to limited accuracy. Targeting towards high accuracy and efficiency, we propose MaKeGNN, a sparse attention-based GNN architecture which bypasses non-repeatable keypoints and leverages matchable ones to guide compact and meaningful message passing. More specifically, our Bilateral Context-Aware Sampling Module first dynamically samples two small sets of well-distributed keypoints with high matchability scores from the image pair. Then, our Matchable Keypoint-Assisted Context Aggregation Module regards sampled informative keypoints as message bottlenecks and thus constrains each keypoint only to retrieve favorable contextual information from intra- and inter- matchable keypoints, evading the interference of irrelevant and redundant connectivity with non-repeatable ones. Furthermore, considering the potential noise in initial keypoints and sampled matchable ones, the MKACA module adopts a matchability-guided attentional aggregation operation for purer data-dependent context propagation. By these means, we achieve the state-of-the-art performance on relative camera estimation, fundamental matrix estimation, and visual localization, while significantly reducing computational and memory complexity compared to typical attentional GNNs.

Elastic Net Constraint-Based Tensor Model for High-Order Graph Matching

Matching technology of point cloud model based on statistical features

Graph Matching Optimization Network for Point Cloud Registration

Spectral Feature Matching Based on Isometric Projection of Matrix

GLMNet: Graph Learning-Matching Networks for Feature Matching

Progressive Probabilistic Graph Matching With Local Consistency Regularization

Image Registration Using Isometric Projection of Graph

Intrinsic Shape Matching Via Tensor-Based Optimization

Learning to Match Features with Discriminative Sparse Graphneuralnetwork

Elastic Net Hypergraph Learning for Image Clustering and Semi-supervised Classification

Graduated Consistency-Regularized Optimization For Multi-Graph Matching

Joint Transformation Learning via the L 2,1 -Norm Metric for Robust Graph Matching

Graph matching based on fast normalized cut and multiplicative update mapping

Tensor Graphical Model: Non-Convex Optimization and Statistical Inference

Learning Feature Matching via Matchable Keypoint-Assisted Graph Neural Network

Ensemble Quadratic Assignment Network for Graph Matching

Geometric Polynomial Constraints in Higher-Order Graph Matching

Deep Graph Matching Based Dense Correspondence Learning Between Non-Rigid Point Clouds.

Spectral Graph Matching and Regularized Quadratic Relaxations I Algorithm and Gaussian Analysis

CSGO: Constrained-Softassign Gradient Optimization For Large Graph Matching

Point Cloud Registration Based on Graph Matching Optimization