AI RESEARCH
Scal3R: Scalable Test-Time Training for Large-Scale 3D Reconstruction
arXiv CS.CV
•
ArXi:2604.08542v1 Announce Type: new This paper addresses the task of large-scale 3D scene reconstruction from long video sequences. Recent feed-forward reconstruction models have shown promising results by directly regressing 3D geometry from RGB images without explicit 3D priors or geometric constraints. However, these methods often struggle to maintain reconstruction accuracy and consistency over long sequences due to limited memory capacity and the inability to effectively capture global contextual cues.