ALoMa paper extends affine correspondence estimation to image-pair level, cutting error to single digits
ducha_aiki · x · 2026-11-17
Researchers from Chalmers and collaborators present ALoMa: Affine Correspondences Beyond Patches, extending affine correspondence (AC) estimation beyond local patches to the image and image-pair levels.
- Instead of estimating affine components separately, an transformer-based affine head jointly learns the full 4-DoF matrix from LoMa (LightGlue-family) keypoint descriptors, supervised directly by depth- and pose-derived ground truth.
- Results: affine error on the Belém Tower example drops from 51.0 (AffNet) and 55.3 (Sun et al.) to 9.4; essential matrix estimation hits AUC@5° of 64.8–66.0 on Mega1500 and 30.0 on ScanNet1500, beating KN+HN+AN and SIFT baselines.
- Consistent gains across rolling shutter correction, image registration and other multi-view geometry tasks; code released open source.
Related event: ALoMa Paper Extends Affine Correspondence Estimation Beyond Patches(2 posts)→
More from Research
- EACL 2027 Paper Finds up to 74% Logical Error Rate in LLM Code Across 10 Models — ponguru · 2026-11-17
- AI agents rewrote 30,000 articles of French law; blind audits found 54% fewer legal inconsistencies — IgorCarron · 2026-11-17
- COrigami: open-source pipeline turns prompts into flat-foldable origami with Gemini and RL — TZahavy · 2026-11-17
- How rollout staleness hurts on-policy distillation, and how to build a faster pipeline — burkov · 2026-11-17
- Apple robotics paper: 91% success in unfamiliar homes using synthetic sim data — FuSheng_0306 · 2026-11-17
- Hierarchos-Native v0.2.0: Rust+Vulkan LLM training adds RDNA3 WMMA and native multimodal training — PhysicsDisastrous462 · 2026-11-17