LATIDIA · Robótica

ScaRF-SLAM: Reconstrucción consistente a escala con modelos de avance y SLAM visual clásico

arXiv:2606.00307v4 Announce Type: replace Resumen: Trabajos recientes han explorado la unificación de SLAM con modelos de cimentación geométrica (GFM). Sin embargo, el uso directo de las predicciones de GFM para el seguimiento es muy sensible a la capa del modelo

La noticia

arXiv:2606.00307v4 Announce Type: replace Abstract: Recent works have explored unifying SLAM with geometric foundation models (GFMs). However, directly using GFM predictions for tracking is highly sensitive to model capability and uncertainty, as geometric inaccuracies in the predictions can adversely affect pose estimation. To address this limitation, we propose a decoupled framework that integrates classical feature-based SLAM with GFMs, which achieves higher quality and more consistent dense reconstruction. In brief, we use classical visual SLAM for robust low-latency tracking and use GFMs exclusively for mapping. By anchoring mapping to poses produced by the SLAM module and optimizing across depth scales, the proposed design avoids propagating inaccuracies from GFM predictions into pose