Compositional Scalable Object SLAM
This addresses the challenge of robust SLAM for artificial environments, offering a solution for robotics and AR/VR applications, though it appears incremental by building on object-based mapping.
The paper tackles the problem of drift-free large-scale indoor reconstruction in SLAM by representing scenes as a graph of objects, achieving fast, scalable, and accurate performance with an online implementation that runs at near frame rate.
We present a fast, scalable, and accurate Simultaneous Localization and Mapping (SLAM) system that represents indoor scenes as a graph of objects. Leveraging the observation that artificial environments are structured and occupied by recognizable objects, we show that a compositional scalable object mapping formulation is amenable to a robust SLAM solution for drift-free large scale indoor reconstruction. To achieve this, we propose a novel semantically assisted data association strategy that obtains unambiguous persistent object landmarks, and a 2.5D compositional rendering method that enables reliable frame-to-model RGB-D tracking. Consequently, we deliver an optimized online implementation that can run at near frame rate with a single graphics card, and provide a comprehensive evaluation against state of the art baselines. An open source implementation will be provided at https://placeholder.