BS3D: Building-scale 3D Reconstruction from RGB-D Images
This provides a crowd-sourced, cost-effective dataset for researchers in SLAM and 3D reconstruction, though it is incremental as it builds on existing methods with new data.
The authors tackled the problem of limited datasets for SLAM by proposing an easy-to-use framework for building-scale 3D reconstruction using a consumer depth camera, resulting in a new dataset (BS3D) that improved monocular depth estimation models and enabled benchmarking of visual-inertial odometry methods.
Various datasets have been proposed for simultaneous localization and mapping (SLAM) and related problems. Existing datasets often include small environments, have incomplete ground truth, or lack important sensor data, such as depth and infrared images. We propose an easy-to-use framework for acquiring building-scale 3D reconstruction using a consumer depth camera. Unlike complex and expensive acquisition setups, our system enables crowd-sourcing, which can greatly benefit data-hungry algorithms. Compared to similar systems, we utilize raw depth maps for odometry computation and loop closure refinement which results in better reconstructions. We acquire a building-scale 3D dataset (BS3D) and demonstrate its value by training an improved monocular depth estimation model. As a unique experiment, we benchmark visual-inertial odometry methods using both color and active infrared images.