Planar-SfM: Camera Pose Estimation via Homography Graph Embeddings
Gabi Pragier ∗, Matan Karklinsky ∗, David Ungarish ∗, Avi Ben-Cohen · 2026
Read Planar-SfM: Camera Pose Estimation via Homography Graph Embeddings by Gabi Pragier ∗, Matan Karklinsky ∗, David Ungarish ∗, Avi Ben-Cohen (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-31979.txt.
Description
Pragier and three others turn planar surfaces from a weakness of structure-from-motion into a source of constraints. Standard pose estimation rests on epipolar geometry, which degenerates when a scene is dominated by flat surfaces like indoor sports courts. Each plane seen across views, though, yields its own estimate of relative camera pose through homography decomposition.
Planar-SfM builds a pose graph from those homography estimates. Spectral embedding scores the edges by geometric and visual consistency, filtering unreliable ones before a maximally consistent spanning tree is extracted for pose recovery. On basketball court imagery, where current methods struggle, the approach performs better. On the unconstrained IMC Phototourism benchmark it matches or exceeds state-of-the-art results.
Catalog record
| CREATOR | Gabi Pragier ∗, Matan Karklinsky ∗, David Ungarish ∗, Avi Ben-Cohen |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 21 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2606.31979