AI Paper Atlas

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

Towards Improved and Interpretable Deep Metric Learning via Attentive Grouping

Xinyi Xu, Zhengyang Wang, Cheng Dan Deng, Hao Yuan, Shuiwang Ji

TPAMI 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 19 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 A-grouping (4x128) instead of no grouping (ablation)ResNet-50 + Margin · Cars-196Recall@1: 79.92 -> 87.06Improved
add A-grouping (4x128) instead of no grouping (ablation)ResNet-50 + Margin · CUB-200-2011Recall@1: 63.09 -> 68.62Improved
add A-grouping (4x128) instead of no grouping (ablation)Inception-v1 + Binomial · Cars-196Recall@1: 71.70 -> 80.00Improved
add A-grouping (4x128) instead of no grouping (ablation)Inception-v1 + Margin · Cars-196Recall@1: 73.12 -> 80.23Improved
add A-grouping (4x128) instead of no grouping (ablation)Inception-v1 + Contrastive · CUB-200-2011Recall@1: 54.74 -> 57.06Improved
add A-grouping (4x128) instead of no grouping (ablation)ResNet-50 + Binomial · Cars-196Recall@1: 80.90 -> 85.70Improved
add A-grouping (4x128) instead of no grouping (ablation)ResNet-50 + Margin · SOPRecall@1: 78.32 -> 78.70No clear effect
use A-grouping instead of channel grouping (ablation)ResNet-50 + Margin · Cars-196Recall@1: 84.93 -> 87.06Improved
use A-grouping instead of channel grouping (ablation)Inception-v1 + Contrastive · Cars-196Recall@1: 71.84 -> 71.34Hurt
use A-grouping instead of channel grouping (ablation)ResNet-50 + Margin · SOPRecall@1: 77.66 -> 78.70Improved
use gate grouping instead of A-grouping (ablation)ResNet-50 + Margin · SOPRecall@1: 78.70 -> 76.58Hurt
use multi-linear grouping instead of A-grouping (ablation)ResNet-50 + Binomial · Cars-196Recall@1: 85.70 -> 82.33Hurt

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