GPO: Global Plane Optimization for Fast and Accurate Monocular SLAM Initialization
This addresses faster and more accurate initialization for monocular SLAM systems, though it appears incremental as it builds on existing planar feature methods.
The paper tackles monocular SLAM initialization by proposing a global plane optimization method that uses planar features to estimate camera poses and plane normals without triangulation. Experimental results show it outperforms baselines in accuracy and real-time performance on a chessboard dataset.
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.