ALoMa: Affine Correspondences Beyond Patches
Lists
Only you see this
Abstract
Affine correspondences (ACs) enable robust and efficient solvers for multiple-view geometry by constraining the local geometry of point correspondences (PCs). Despite their utility in minimal solvers, AC estimation remains understudied. We revisit AC estimation beyond local image patches, extending it to the image and image-pair levels. Rather than estimating the components of the affine transformation separately, we learn the full 4-DoF matrix jointly. Building on this formulation, we introduce ALoMa, an AC estimator using LoMa keypoint descriptions, demonstrating consistent improvements across a wide range of downstream tasks. Code at https://github.com/davnords/loma.
BibTeX
@misc{nordstrom2026aloma,
author = {David Nordström and Georg Bökman and Viktor Larsson and Fredrik Kahl and Jonathan Ventura},
title = {{ALoMa: Affine Correspondences Beyond Patches}},
year = {2026},
eprint = {2611.18170},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.18170},
}Loading related content...