AI Paper Atlas

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

Better and Faster: Exponential Loss for Image Patch Matching

Shuang Wang, Yanfeng Li, Xuefeng Liang, Dou Quan, Bowu Yang, Shaowei Wei, Licheng Jiao

ICCV 2019 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 28 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 linear triplet loss with exponential triplet loss (metric learning) (ablation)HardNet-style network · UBCmean FPR95: 1.56 -> 0.98Improved
replace linear triplet loss with exponential triplet loss (descriptor learning) (ablation)HardNet · UBCmean FPR95: 1.25 -> 1.00Improved
replace linear with exponential losses in HardNet (ablation)HardNet · UBCtraining timeImproved
add exponential Siamese loss instead of linear on MNIST toy (ablation)Siamese network · MNIST clusteringSTD of intra-class distances: 0.71 -> 0.57Improved
add exponential triplet loss instead of linear on MNIST toy (ablation)triplet network · MNIST clusteringSTD of intra-class distances: 0.81 -> 0.65Improved
add hard positive mining at ratio 1:2 (batch size 128) (ablation)Exp-TLoss (dist) · UBCFPR95: 1.09 -> 1.03Improved
use exponential orders beta=2, gamma=0.3 rather than linear orders (ablation)Exp-TLoss (prob) · UBC (Liberty-trained)FPR95: 0.75Improved
train on full UBC dataset (FB) instead of Liberty onlyExp-TLoss (dist) · HPatches patch verificationmAP: 87.70% -> 90.07%Improved
use Exp-TLoss (dist) instead of HardNet+ when trained on Libertydescriptor learning · HPatches patch verificationmAP: 88.43% -> 87.70%Hurt
use Exp-TLoss (dist) instead of HardNet+ on RGB-NIRdescriptor learning · RGB-NIRmean FPR95: 2.80 -> 2.07Improved
use Exp-TLoss (prob) instead of SNet on Notredame-trained UBCmetric learning · UBCFPR95Mixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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