CLoSeR builds on LoGeR, a streaming backbone that predicts camera poses, depth maps, and dense point clouds across windows with globally consistent scale.
Loop closure in streaming 3D reconstruction achieves kilometer-scale accuracy without drift
By exploiting the globally consistent scale of the streaming backbone LoGeR, the method recovers relative poses via loop-conditioned windows and optimizes on the SE(3) manifold.
Big Tech
Moyang Li · Zihan Zhu · Wei Zhang · Marc Pollefeys · Daniel Barath
ETH Zurich · University of Stuttgart · Microsoft
Research Digest··3 min read
The authors present CLoSeR, a system that integrates loop closure into a feedforward streaming reconstruction backbone (LoGeR) to eliminate tracking drift on long trajectories.
Why this paper
From Microsoft and 2 others
In one line
CLoSeR adds loop closure to a streaming reconstruction backbone to achieve drift-free kilometer-scale 3D reconstruction from monocular video.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ✓Limitations stated by the authors
- ·No benchmark numbers found
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§