AI Paper Atlas

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

Learning Spread-Out Local Feature Descriptors

Xu Zhang, Felix X. Yu, Sanjiv Kumar, Shih-Fu Chang

ICCV 2017 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 142 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 global orthogonal regularization (GOR) to contractive loss, versus best pairwise model DeepCompare 2strcontractive loss · UBC patch mean FPR95FPR95 (%): 9.67 -> 5.97Improved
add GOR to triplet lossTFeat triplet loss · UBC patch mean FPR95FPR95 (%): 6.79 -> 4.69Improved
add GOR to triplet loss with anchor swapTFeat+AS · UBC patch mean FPR95FPR95 (%): 6.47 -> 4.36Improved
add GOR to N-pair lossN-pair · UBC patch mean FPR95FPR95 (%): 5.29 -> 4.98Improved
add anchor swap on top of triplet loss with GORTL+GOR · UBC patch mean FPR95FPR95 (%): 4.69 -> 4.36Improved
use hinge on the second-moment term of GOR, versus l1 or l2 (ablation)GOR · UBC patchFPR95 (%)No clear effect
use small descriptor dimension d=32 with GOR (ablation)GOR with triplet loss · UBC patch, embedding dimensionFPR95 (%)Hurt
use GOR with descriptor dimension d >= 64 (ablation)GOR with triplet loss · UBC patch, embedding dimensionFPR95 (%)Improved
add GOR to LSSS loss for image embeddingLSSS · Stanford Online ProductsF1, NMI, Recall@KImproved
add GOR to metric learning on Car196 and CUB-200-2011embedding loss · Car196 and CUB-200-2011RecallNo clear effect
add GOR, measured by overlap of similarity histogramstriplet loss · UBC Notre Dame -> Libertyempirical error (histogram overlap): 15% relative decreaseImproved

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