Point-SLAM: Dense Neural Point Cloud-based SLAM
This work addresses the challenge of efficient and accurate simultaneous localization and mapping for robotics and AR/VR applications, though it is incremental as it builds on existing dense neural SLAM methods.
The authors tackled the problem of dense neural SLAM for monocular RGBD input by proposing a point cloud-based neural scene representation that adapts anchor density to input information, achieving competitive or better performance in tracking, mapping, and rendering accuracy on datasets like Replica, TUM-RGBD, and ScanNet.
We propose a dense neural simultaneous localization and mapping (SLAM) approach for monocular RGBD input which anchors the features of a neural scene representation in a point cloud that is iteratively generated in an input-dependent data-driven manner. We demonstrate that both tracking and mapping can be performed with the same point-based neural scene representation by minimizing an RGBD-based re-rendering loss. In contrast to recent dense neural SLAM methods which anchor the scene features in a sparse grid, our point-based approach allows dynamically adapting the anchor point density to the information density of the input. This strategy reduces runtime and memory usage in regions with fewer details and dedicates higher point density to resolve fine details. Our approach performs either better or competitive to existing dense neural RGBD SLAM methods in tracking, mapping and rendering accuracy on the Replica, TUM-RGBD and ScanNet datasets. The source code is available at https://github.com/eriksandstroem/Point-SLAM.