ToF-Splatting: Dense SLAM using Sparse Time-of-Flight Depth and Multi-Frame Integration
This addresses the challenge of enabling dense SLAM on mobile and AR/VR devices with power-limited sparse ToF sensors, representing an incremental improvement over existing methods.
The paper tackles the problem of using extremely sparse Time-of-Flight depth data for SLAM by proposing ToF-Splatting, a 3D Gaussian Splatting-based pipeline with multi-frame integration, achieving state-of-the-art tracking and mapping performances on synthetic and real datasets.
Time-of-Flight (ToF) sensors provide efficient active depth sensing at relatively low power budgets; among such designs, only very sparse measurements from low-resolution sensors are considered to meet the increasingly limited power constraints of mobile and AR/VR devices. However, such extreme sparsity levels limit the seamless usage of ToF depth in SLAM. In this work, we propose ToF-Splatting, the first 3D Gaussian Splatting-based SLAM pipeline tailored for using effectively very sparse ToF input data. Our approach improves upon the state of the art by introducing a multi-frame integration module, which produces dense depth maps by merging cues from extremely sparse ToF depth, monocular color, and multi-view geometry. Extensive experiments on both synthetic and real sparse ToF datasets demonstrate the viability of our approach, as it achieves state-of-the-art tracking and mapping performances on reference datasets.