Loopy-SLAM: Dense Neural SLAM with Loop Closures

arXiv:2402.09944v1 Announce Type: new
Abstract: Neural RGBD SLAM techniques have shown promise in dense Simultaneous Localization And Mapping (SLAM), yet face challenges such as error accumulation during camera tracking resulting in distorted maps. In response, we introduce Loopy-SLAM that globally optimizes poses and the dense 3D model. We use frame-to-model tracking using a data-driven point-based submap generation method and trigger loop closures online by performing global place recognition. Robust pose graph optimization is used to rigidly align the local submaps. …

abstract arxiv challenges closures cs.cv error face localization loop mapping maps slam tracking type