Learning to Match Features with Seeded Graph Matching Network

TL;DR

This paper introduces a Seeded Graph Matching Network that leverages seed matches and novel message passing operations to improve accuracy and efficiency in local feature matching across images.

Contribution

It proposes a sparse graph neural network with new message passing operations, reducing complexity while maintaining or improving matching performance.

Findings

Reduces computational and memory complexity significantly.

Achieves competitive or higher matching accuracy.

Introduces novel message passing operations for feature aggregation and propagation.

Abstract

Matching local features across images is a fundamental problem in computer vision. Targeting towards high accuracy and efficiency, we propose Seeded Graph Matching Network, a graph neural network with sparse structure to reduce redundant connectivity and learn compact representation. The network consists of 1) Seeding Module, which initializes the matching by generating a small set of reliable matches as seeds. 2) Seeded Graph Neural Network, which utilizes seed matches to pass messages within/across images and predicts assignment costs. Three novel operations are proposed as basic elements for message passing: 1) Attentional Pooling, which aggregates keypoint features within the image to seed matches. 2) Seed Filtering, which enhances seed features and exchanges messages across images. 3) Attentional Unpooling, which propagates seed features back to original keypoints. Experiments show that our method reduces computational and memory complexity significantly compared with typical attention-based networks while competitive or higher performance is achieved.