# Repeatability Is Not Enough: Learning Affine Regions via Discriminability

arXiv: https://arxiv.org/abs/1711.06704 · HTML: https://mlsanity.com/paper/1711.06704/

## Key facts

- Cited by 57 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Nov 2017.

## Built on by (most cited first, 30 of 57)

- [Grounding Image Matching in 3D with MASt3R](https://mlsanity.com/paper/2406.09756/) (2024)
- [MatchFormer: Interleaving Attention in Transformers for Feature Matching](https://mlsanity.com/paper/2203.09645/) (2022)
- [ALIKED: A Lighter Keypoint and Descriptor Extraction Network via Deformable Transformation](https://mlsanity.com/paper/2304.03608/) (2023)
- [DeDoDe: Detect, Don't Describe -- Describe, Don't Detect for Local Feature Matching](https://mlsanity.com/paper/2308.08479/) (2023)
- [DOLG: Single-Stage Image Retrieval with Deep Orthogonal Fusion of Local and Global Features](https://mlsanity.com/paper/2108.02927/) (2021)
- [Correlation Verification for Image Retrieval](https://mlsanity.com/paper/2204.01458/) (2022)
- [Iterative Deep Homography Estimation](https://mlsanity.com/paper/2203.15982/) (2022)
- [LocalTrans: A Multiscale Local Transformer Network for Cross-Resolution Homography Estimation](https://mlsanity.com/paper/2106.04067/) (2021)
- [Decoupling Makes Weakly Supervised Local Feature Better](https://mlsanity.com/paper/2201.02861/) (2022)
- [SFD2: Semantic-guided Feature Detection and Description](https://mlsanity.com/paper/2304.14845/) (2023)
- [Self-Supervised Equivariant Learning for Oriented Keypoint Detection](https://mlsanity.com/paper/2204.08613/) (2022)
- [Local Feature Matching Using Deep Learning: A Survey](https://mlsanity.com/paper/2401.17592/) (2024)
- [Learning Rotation-Equivariant Features for Visual Correspondence](https://mlsanity.com/paper/2303.15472/) (2023)
- [SealID: Saimaa ringed seal re-identification dataset](https://mlsanity.com/paper/2206.02260/) (2022)
- [DFM: A Performance Baseline for Deep Feature Matching](https://mlsanity.com/paper/2106.07791/) (2021)
- [A Large Scale Homography Benchmark](https://mlsanity.com/paper/2302.09997/) (2023)
- [Segment Anything Model is a Good Teacher for Local Feature Learning](https://mlsanity.com/paper/2309.16992/) (2023)
- [Learning to Make Keypoints Sub-Pixel Accurate](https://mlsanity.com/paper/2407.11668/) (2024)
- [ScaleNet: A Shallow Architecture for Scale Estimation](https://mlsanity.com/paper/2112.04846/) (2021)
- [Two-View Geometry Scoring Without Correspondences](https://mlsanity.com/paper/2306.01596/) (2023)
- [LoMa: Local Feature Matching Revisited](https://mlsanity.com/paper/2604.04931/) (2026)
- [SIFT Matching by Context Exposed](https://mlsanity.com/paper/2106.09584/) (2021)
- [Self-Supervised Learning of Image Scale and Orientation](https://mlsanity.com/paper/2206.07259/) (2022)
- [DenseGAP: Graph-Structured Dense Correspondence Learning with Anchor Points](https://mlsanity.com/paper/2112.06910/) (2021)
- [LFM-3D: Learnable Feature Matching Across Wide Baselines Using 3D Signals](https://mlsanity.com/paper/2303.12779/) (2023)
- [Deep Learning Reforms Image Matching: A Survey and Outlook](https://mlsanity.com/paper/2506.04619/) (2025)
- [SuperPoint features in endoscopy](https://mlsanity.com/paper/2203.04302/) (2022)
- [Relative Pose from SIFT Features](https://mlsanity.com/paper/2203.07930/) (2022)
- [Handbook on Leveraging Lines for Two-View Relative Pose Estimation](https://mlsanity.com/paper/2309.16040/) (2023)
- [Affine steerers for structured keypoint description](https://mlsanity.com/paper/2408.14186/) (2024)

## Abstract (arXiv)

A method for learning local affine-covariant regions is presented. We show that maximizing geometric repeatability does not lead to local regions, a.k.a features,that are reliably matched and this necessitates descriptor-based learning. We explore factors that influence such learning and registration: the loss function, descriptor type, geometric parametrization and the trade-off between matchability and geometric accuracy and propose a novel hard negative-constant loss function for learning of affine regions. The affine shape estimator -- AffNet -- trained with the hard negative-constant loss outperforms the state-of-the-art in bag-of-words image retrieval and wide baseline stereo. The proposed training process does not require precisely geometrically aligned patches.The source codes and trained weights are available at https://github.com/ducha-aiki/affnet