Uploaded June 2026 | Updated September 2026, 3 weeks ago
ScaRF-SLAM introduces a decoupled framework that integrates classical feature-based SLAM with GFMs, which achieves higher quality and more consistent dense reconstruction. Experiments show that our approach achieves superior trajectory accuracy while improving reconstruction precision by 10%-20% over existing methods, with about 2 cm reconstruction error per 10 m chunk on building-scale dataset. On large-scale outdoor datasets, it attains 10 cm error per 30 m chunk (w.r.t LiDAR ground-truth models).
Authors: Yuhao Zhang, Yifu Tao, Frank Dellaert, Maurice Fallon
Code and Project Page: github.com/ori-drs/ScaRF-SLAM
Paper: arxiv.org/abs/2606.00307v1
ScaRF-SLAM introduces a decoupled framework that integrates classical feature-based SLAM with GFMs, which achieves higher quality and more consistent dense reconstruction. Experiments show that our approach achieves superior trajectory accuracy while improving reconstruction precision by 10%-20% over existing methods, with about 2 cm reconstruction error per 10 m chunk on building-scale dataset. On large-scale outdoor datasets, it attains 10 cm error per 30 m chunk (w.r.t LiDAR ground-truth models).
Authors: Yuhao Zhang, Yifu Tao, Frank Dellaert, Maurice Fallon
Code and Project Page: github.com/ori-drs/ScaRF-SLAM
Paper: arxiv.org/abs/2606.00307v1
