GPO: Global Plane Optimization for Fast and Accurate Monocular SLAM Initialization

TL;DR

This paper introduces GPO, a novel global plane optimization method for monocular SLAM initialization that leverages planar features for fast, accurate pose estimation without triangulation, outperforming baseline methods.

Contribution

The paper presents a new initialization approach for monocular SLAM using global plane optimization that improves accuracy and speed by exploiting planar constraints across multiple frames.

Findings

Outperforms baseline methods in accuracy

Operates in real-time

Effectively exploits planar features for pose estimation

Abstract

Initialization is essential to monocular Simultaneous Localization and Mapping (SLAM) problems. This paper focuses on a novel initialization method for monocular SLAM based on planar features. The algorithm starts by homography estimation in a sliding window. It then proceeds to a global plane optimization (GPO) to obtain camera poses and the plane normal. 3D points can be recovered using planar constraints without triangulation. The proposed method fully exploits the plane information from multiple frames and avoids the ambiguities in homography decomposition. We validate our algorithm on the collected chessboard dataset against baseline implementations and present extensive analysis. Experimental results show that our method outperforms the fine-tuned baselines in both accuracy and real-time.