AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Learning With Average Precision: Training Image Retrieval With a Listwise Loss

Jerome Revaud, Jon Almazan, Rafael S. Rezende, Cesar Roberto de Souza

ICCV 2019 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 426 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 triplet loss with hard negative mining by the listwise mAP loss (ablation)GeM ResNet-101 · ROxford MediummAP: 64.9 -> 67.4Improved
replace triplet loss with hard negative mining by the listwise mAP loss (ablation)GeM ResNet-101 · RParis HardmAP: 58.7 -> 61.0Improved
increase triplet batch from 64 to 1024 triplets (ablation)GeM (TL) · ROxford MediummAP: 64.9 -> 65.5No clear effect
add class weighting inside each batch (ablation)GeM (AP) · Revisited Oxford/Paris MediummAP: +2%Improved
use tie-aware AP formulation instead of plain quantized AP (ablation)GeM (AP) · Revisited Oxford/Paris MediummAPHurt
increase batch size up to 4096 (ablation)GeM (AP) · Revisited Oxford/Paris MediummAP: saturates beyond 4096Improved
sample batch from a limited random set of classes (ablation)GeM (AP) · Revisited Oxford/Paris MediummAPNo clear effect
vary number of quantization bins M (ablation)GeM (AP) · Revisited Oxford/Paris MediummAP: M=20 usedNo clear effect
add multi-scale testing (ablation)GeM (AP) · Revisited Oxford/ParismAPNo clear effect
add weighted query expansionGeM (AP) · ROxford MediummAP: 67.5 -> 71.4Improved
fine-tune with mAP loss instead of off-the-shelf featuresGeM ResNet-101 · ROxford MediummAP: 45.0 -> 67.5Improved
use listwise mAP loss instead of triplet loss for training cost (ablation)GeM ResNet-101 · Training costtraining time: 3 days -> 1 dayImproved

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