MambaGlue: Fast and Robust Local Feature Matching With Mamba

TL;DR

MambaGlue introduces a fast, robust local feature matching method leveraging Mamba architecture, combining local-global context understanding and confidence scoring to outperform baselines in efficiency and accuracy.

Contribution

The paper presents MambaGlue, a novel local feature matching approach that integrates Mamba-based self-attention and confidence scoring for improved speed and robustness.

Findings

Significant performance improvement over baseline methods.

Maintains fast inference speed in real-world datasets.

Balances robustness and efficiency effectively.

Abstract

In recent years, robust matching methods using deep learning-based approaches have been actively studied and improved in computer vision tasks. However, there remains a persistent demand for both robust and fast matching techniques. To address this, we propose a novel Mamba-based local feature matching approach, called MambaGlue, where Mamba is an emerging state-of-the-art architecture rapidly gaining recognition for its superior speed in both training and inference, and promising performance compared with Transformer architectures. In particular, we propose two modules: a) MambaAttention mixer to simultaneously and selectively understand the local and global context through the Mamba-based self-attention structure and b) deep confidence score regressor, which is a multi-layer perceptron (MLP)-based architecture that evaluates a score indicating how confidently matching predictions correspond to the ground-truth correspondences. Consequently, our MambaGlue achieves a balance between robustness and efficiency in real-world applications. As verified on various public datasets, we demonstrate that our MambaGlue yields a substantial performance improvement over baseline approaches while maintaining fast inference speed. Our code will be available on https://github.com/url-kaist/MambaGlue