Search papers, labs, and topics across Lattice.
The Infinite SLAM Transformer (SLAMFormer-$\infty$) introduces a novel geometric transformer that facilitates unbounded frontend and backend processing by utilizing memory conditions to create flexible coordinate systems. This approach allows for efficient local computations in the frontend while enabling the backend to optimize long-range trajectories and scene geometry with global consistency. Experimental results indicate that SLAMFormer-$\infty$ outperforms or matches existing methods in trajectory estimation and scene reconstruction, even for sequences exceeding 17 km in length.
SLAMFormer-$\infty$ can handle trajectory sequences over 17 km without losing performance, revolutionizing long-range SLAM capabilities.
We introduce the Infinite SLAM Transformer (SLAMFormer-$\infty$), the first geometric transformer capable of supporting both long-range frontend and backend processing without an explicit distance bound. Instead of relying on a first-frame-anchored formulation, SLAMFormer-$\infty$ employs memory conditions to define flexible coordinate systems and scales for input frames, enabling more expressive structural conditioning. Built upon this formulation, the frontend preserves efficient local computation, while the backend jointly optimizes long-range trajectories and scene geometry in a globally consistent manner. Experimental results demonstrate that SLAMFormer-$\infty$ achieves superior or highly competitive performance in both trajectory estimation and scene reconstruction across large-scale datasets. Notably, SLAMFormer-$\infty$ generalizes to extremely long trajectories, successfully operating on sequences exceeding $17\mathrm{km}$.
