AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

CNN Image Retrieval Learns from BoW: Unsupervised Fine-Tuning with Hard Examples

Filip Radenović, Giorgos Tolias, Ondřej Chum

ECCV 2016 · Architectures and efficiency › Pruning and efficient tuning · approach: Efficient ViT adaptation · 565 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
fine-tune the CNN on SfM-derived pairs instead of off-the-shelfVGG MAC 512D · Oxford5kmAP: 56.4 -> 79.7Improved
add learned whitening Lw instead of no post-processing (ablation)fine-tuned VGG MAC · Oxford5kmAP: 78.7 -> 79.7Improved
use learned whitening Lw instead of PCA whitening (ablation)fine-tuned VGG MAC · Oxford5kmAP: 76.1 -> 79.7Improved
add PCA whitening instead of no post-processing (ablation)fine-tuned AlexNet MAC · Oxford105kmAP: 54.2 -> 44.1Hurt
add PCA whitening instead of no post-processing (ablation)fine-tuned VGG MAC · Oxford5k / Paris6kmAP: 78.7 -> 76.1 (Oxford5k); 77.1 -> 79.0 (Paris6k)Mixed
select positives by relaxed inliers (m3) instead of MAC distance (m1) (ablation)siamese MAC · Oxford105k / Paris106kmAPImproved
select negatives with at most one per cluster (N2) instead of kNN (N1) (ablation)siamese MAC · Oxford105k / Paris106kmAPImproved
use contrastive loss instead of triplet loss (ablation)siamese MAC · Oxford105k / Paris106kmAPImproved
train with all 551 3D models instead of 10 or 100 (ablation)AlexNet MAC · Oxford105k / Paris106kmAPImproved
include Oxford and Paris landmarks in the training reconstructions (ablation)fine-tuned MAC · Oxford/ParismAP: +0.3No clear effect
replace MAC with R-MAC regional aggregationfine-tuned VGG 512D · Oxford5k / Paris6kmAP: 79.7 -> 77.0 (Oxford5k); 82.4 -> 83.8 (Paris6k)Mixed
add re-ranking and query expansion to MACfine-tuned VGG MAC · Oxford5kmAP: 79.7 -> 85.0Improved

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