An Error-Matching Exclusion Method for Accelerating Visual SLAM

TL;DR

This paper introduces a faster Visual SLAM feature matching method by combining GMS and RANSAC, significantly reducing runtime while maintaining accuracy across multiple datasets.

Contribution

It presents a novel integration of GMS and RANSAC with a confidence-based sample prioritization to accelerate feature matching in Visual SLAM.

Findings

Reduces average runtime by 24.13% on benchmark datasets

Maintains comparable accuracy to traditional GMS-RANSAC

Effective in real-time Visual SLAM applications

Abstract

In Visual SLAM, achieving accurate feature matching consumes a significant amount of time, severely impacting the real-time performance of the system. This paper proposes an accelerated method for Visual SLAM by integrating GMS (Grid-based Motion Statistics) with RANSAC (Random Sample Consensus) for the removal of mismatched features. The approach first utilizes the GMS algorithm to estimate the quantity of matched pairs within the neighborhood and ranks the matches based on their confidence. Subsequently, the Random Sample Consensus (RANSAC) algorithm is employed to further eliminate mismatched features. To address the time-consuming issue of randomly selecting all matched pairs, this method transforms it into the problem of prioritizing sample selection from high-confidence matches. This enables the iterative solution of the optimal model. Experimental results demonstrate that the proposed method achieves a comparable accuracy to the original GMS-RANSAC while reducing the average runtime by 24.13% on the KITTI, TUM desk, and TUM doll datasets.