Repeatability Is Not Enough: Learning Affine Regions via Discriminability

ECCV 2018  ·  Dmytro Mishkin, Filip Radenovic, Jiri Matas ·

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

PDF Abstract ECCV 2018 PDF ECCV 2018 Abstract

Results from the Paper

Ranked #4 on Image Matching on IMC PhotoTourism (using extra training data)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Result Benchmark
Image Matching IMC PhotoTourism DoG-AffNet-HardNet8 mean average accuracy @ 10 0.64212 # 4


No methods listed for this paper. Add relevant methods here