AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Fine-Tuning CNN Image Retrieval with No Human Annotation

Filip Radenović, Giorgos Tolias, Ondřej Chum

TPAMI 2018 · Architectures and efficiency › Pruning and efficient tuning · approach: Efficient ViT adaptation · 1,587 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
replace max pooling (MAC) with learned generalized-mean pooling (GeM)fine-tuned AlexNet · Oxford5kmAP: 62.2 -> 67.7Improved
replace average pooling (SPoC) with learned GeMfine-tuned AlexNet · Paris6kmAP: 70.8 -> 75.5Improved
learn one pooling parameter per feature map instead of a single shared p (ablation)GeM AlexNet · Oxford5kmAP: 67.7 -> 66.3Hurt
learn the shared GeM parameter instead of fixing p=3 (ablation)GeM AlexNet · Paris6kmAP: 74.8 -> 75.5Improved
add learned whitening Lw instead of PCA whitening (ablation)fine-tuned VGG GeM · Paris6kmAP: 84.5 -> 86.0Improved
add PCA whitening to raw descriptors (ablation)fine-tuned VGG MAC/GeM · Oxford5k (AlexNet MAC)mAP: 62.2 -> 56.9Hurt
learn whitening end-to-end inside the network instead of post-hoc Lw (ablation)AlexNet MAC · Oxford105kmAP: 52.8 -> 49.6Hurt
replace average pooling of multi-scale descriptors with GeM (ablation)VGG GeM, Lw · Oxford5kmAP: 87.2 -> 87.9Improved
add multi-scale representation (scales 1, 1/sqrt2, 1/2) at test time (ablation)VGG GeM, Lw · Oxford5kmAP: 85.9 -> 87.9Improved
add alpha-weighted query expansionVGG GeM multi-scale Lw · Oxford5kmAP: 87.9 -> 91.9Improved
train with models containing Oxford and Paris landmarks (ablation)fine-tuned network · Oxford5k and Paris6kmAP: +0.3No clear effect
reduce descriptor dimensionality from 512 to 128 with Lw (ablation)VGG GeM multi-scale · Oxford5kmAP: 87.9 -> 81.6Hurt

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