AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Understanding and Improving Kernel Local Descriptors

Arun Mukundan, Giorgos Tolias, Andrei Bursuc, Hervé Jeǵou, Ondřej Chum

IJCV 2018 · Architectures and efficiency › CNN design · approach: Dense matching and estimation CNNs · 11 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add Cartesian parametrization to polarpolar descriptor · Photo TourismFPR95 (%): 22.42 -> 25.37Hurt
add supervised whitening W_S to polar+cartes (ablation)polar + cartes · Photo TourismFPR95 (%): 25.37 -> 5.94Improved
add Cartesian parametrization to polar under supervised whitening (ablation)polar + W_S · Photo TourismFPR95 (%): 7.06 -> 5.94Improved
replace PCA+SQRT with learned supervised whitening (ablation)polar · Photo TourismFPR95 (%): 8.30 -> 7.06Improved
use unsupervised attenuated whitening instead of supervisedpolar + cartes · Photo TourismFPR95 (%): 5.94 -> 6.79Hurt
use shrinkage whitening W_US instead of attenuated W_UA (ablation)polar + cartes · Photo TourismFPR95 (%): 6.79 -> 7.22No clear effect
add unsupervised whitening W_US (learned on PT) to raw descriptor (ablation)polar + cartes · HPatches retrievalmAP (%): 45.23 -> 53.50Improved
use unsupervised instead of supervised whitening when learned on PT (ablation)polar + cartes · HPatches retrievalmAP (%): 49.66 -> 53.50Improved
learn supervised whitening on HP instead of PT (ablation)polar + cartes + W_S · HPatches retrievalmAP (%): 49.66 -> 61.79Improved
learn unsupervised W_UA on HP instead of PT (ablation)polar + cartes + W_UA · HPatches retrievalmAP (%): 52.78 -> 56.36Improved
learn unsupervised W_US on HP instead of PT (ablation)polar + cartes + W_US · HPatches verificationmAP (%): 78.81 -> 80.70Improved
vary shrinkage parameter t or beta (ablation)W_UA / W_US · Photo Tourism and HPatchesFPR95 (%)No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy