LightGlueStick: a Fast and Robust Glue for Joint Point-Line Matching

2025-10-18

AI Summary

Traditional point and line matching are typically treated as separate tasks; while existing joint approaches (e.g., GlueStick) reduce computational overhead, their architectural complexity hinders real-time deployment. This paper proposes a lightweight end-to-end joint matching framework centered on an Attention-based Line Message Passing (ALMP) mechanism, which explicitly encodes segment topological connectivity and integrates point/line geometric priors into a heterogeneous graph neural network. The method achieves state-of-the-art performance on HPatches, ETH3D, and other benchmarks—significantly outperforming both unimodal and mainstream joint matching methods—while maintaining extremely low parameter count (<1M) and inference latency (<10 ms on GPU). Its efficiency enables direct deployment on embedded devices, making it suitable for real-time visual geometry applications such as SLAM and SfM.

Technology Category

• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Hardware-aware ML

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Lines and points are complementary local features, whose combination has proven effective for applications such as SLAM and Structure-from-Motion. The backbone of these pipelines are the local feature matchers, establishing correspondences across images. Traditionally, point and line matching have been treated as independent tasks. Recently, GlueStick proposed a GNN-based network that simultaneously operates on points and lines to establish matches. While running a single joint matching reduced the overall computational complexity, the heavy architecture prevented real-time applications or deployment to edge devices. Inspired by recent progress in point matching, we propose LightGlueStick, a lightweight matcher for points and line segments. The key novel component in our architecture is the Attentional Line Message Passing (ALMP), which explicitly exposes the connectivity of the lines to the network, allowing for efficient communication between nodes. In thorough experiments we show that LightGlueStick establishes a new state-of-the-art across different benchmarks. The code is available at https://github.com/aubingazhib/LightGlueStick.

Problem

• Develops lightweight joint point-line matching for real-time applications
• Addresses computational complexity of heavy GNN-based feature matchers
• Enables efficient point-line correspondence on edge devices

Innovation

• Lightweight matcher for points and line segments
• Attentional Line Message Passing for connectivity
• Efficient communication between nodes in network