AI Paper Atlas

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

Towards Interpretable Deep Metric Learning With Structural Matching

Wenliang Zhao, Yongming Rao, Ziyi Wang, Jiwen Lu, Jie Zhou

ICCV 2021 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 46 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 DIML structural matching at test time (no training)ProxyAnchor · Cars196Precision@1: 82.36 -> 86.13Improved
add DIML at test timeMargin · CUB200-2011Precision@1: 62.47 -> 65.16Improved
add DIML at test timeContrastive · Cars196Precision@1: 67.45 -> 73.35Improved
add DIML at test timeProxyAnchor · SOPPrecision@1: 79.10 -> 79.22No clear effect
add multi-scale matching to structural similarity (ablation)Margin + DIML · SOPPrecision@1: 77.30 -> 78.53Improved
add structural similarity alone replacing cosine similarity (ablation)Margin · SOPPrecision@1: 78.39 -> 77.30Hurt
add structural similarity alone replacing cosine similarity (ablation)MS · SOPPrecision@1: 77.90 -> 75.81Hurt
add structural similarity alone replacing cosine similarity (ablation)Margin · Cars196Precision@1: 72.18 -> 74.86Improved
add cross-correlation marginals (ablation)Margin + DIML (SS+MM) · Cars196Precision@1: 76.02 -> 76.62Improved
add cross-correlation marginals (ablation)MS + DIML (SS+MM) · SOPPrecision@1: 77.87 -> 78.53Improved
increase truncation number K from 10 to 100 (ablation)Margin + DIML · CUB200-2011Precision@1: 65.16 -> 65.16No clear effect
increase feature map size G from 1 to 7 (ablation)Margin + DIML · Cars196Precision@1: 72.18 -> 76.96Improved

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