MIPS-Fusion: Multi-Implicit-Submaps for Scalable and Robust Online Neural RGB-D Reconstruction
This work addresses the challenge of flexible and scalable 3D reconstruction for robotics or AR/VR applications, though it appears incremental as it builds on existing neural implicit methods.
The authors tackled the problem of scalable and robust online neural RGB-D reconstruction by introducing MIPS-Fusion, which uses a multi-implicit-submap representation to achieve higher reconstruction quality than state-of-the-art methods for large-scale scenes and under fast camera motions.
We introduce MIPS-Fusion, a robust and scalable online RGB-D reconstruction method based on a novel neural implicit representation -- multi-implicit-submap. Different from existing neural RGB-D reconstruction methods lacking either flexibility with a single neural map or scalability due to extra storage of feature grids, we propose a pure neural representation tackling both difficulties with a divide-and-conquer design. In our method, neural submaps are incrementally allocated alongside the scanning trajectory and efficiently learned with local neural bundle adjustments. The submaps can be refined individually in a back-end optimization and optimized jointly to realize submap-level loop closure. Meanwhile, we propose a hybrid tracking approach combining randomized and gradient-based pose optimizations. For the first time, randomized optimization is made possible in neural tracking with several key designs to the learning process, enabling efficient and robust tracking even under fast camera motions. The extensive evaluation demonstrates that our method attains higher reconstruction quality than the state of the arts for large-scale scenes and under fast camera motions.