CLoSeR: Closing the Loop for Long-Context Streaming Reconstruction

2026-10-01

AI Summary

This study addresses the trajectory drift problem in long-context streaming 3D reconstruction with feedforward foundation models, which arises from accumulated errors. We propose a drift-free reconstruction framework based on loop closure detection. The method retrieves candidate frames via global descriptors and constructs conditional windows to estimate relative poses. By leveraging the globally consistent scale of the backbone network, it optimizes poses directly on the SE(3) manifold, thereby avoiding the high-dimensional optimization overhead associated with traditional Sim(3) or SL(4) representations. Experimental results demonstrate that our approach significantly suppresses cumulative drift over kilometer-scale long sequences, generating geometrically consistent, high-quality scenes while substantially outperforming existing state-of-the-art methods.

Abstract

Feedforward foundation models have recently shown remarkable 3D reconstruction capabilities. However, existing models exhibit large tracking drift in long-context streaming reconstruction due to error accumulation. In this paper, we revisit loop closure with streaming reconstruction foundation models to enable accurate, drift-free, kilometer-scale reconstruction. Specifically, our method detects loop candidates through global descriptor retrieval, and constructs loop-conditioned windows to estimate the relative poses between looped frames. Given the observation that our adopted streaming reconstruction backbone produces a globally consistent scale, we optimize all frame poses on the SE(3) manifold with sequential and loop closure constraints, avoiding the pose graph optimization on the Sim(3) or higher-dimensional SL(4) manifolds employed in prior works. Extensive experiments show that our method reduces drift and produces consistent geometry on kilometer-scale sequences, significantly outperforming the state of the art. Code is available at https://github.com/MoyangLi00/CLoSeR.git.

Problem

• 3D reconstruction
• long-context streaming
• tracking drift
• error accumulation
• loop closure

Innovation

• Loop Closure
• Streaming 3D Reconstruction
• Long-Context
• SE(3) Optimization
• Drift-Free