AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Unsupervised Deep Metric Learning with Transformed Attention Consistency and Contrastive Clustering Loss

Yang Li, Shichao Kan, Zhihai He

ECCV 2020 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Pretext-task visual pre-training · 25 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 contrastive clustering loss (ablation)TAC-CCL baseline · CUB-200-2011Recall@1 (%): 53.9 -> 55.7Improved
add contrastive clustering loss (ablation)TAC-CCL baseline · Cars-196Recall@1 (%): 43.0 -> 44.7Improved
add contrastive clustering loss (ablation)TAC-CCL baseline · Stanford Online ProductsRecall@1 (%): 62.5 -> 63.0Improved
add transformed attention consistency on top of CCL (ablation)TAC-CCL · CUB-200-2011Recall@1 (%): 55.7 -> 57.5Improved
add transformed attention consistency on top of CCL (ablation)TAC-CCL · Cars-196Recall@1 (%): 44.7 -> 46.1Improved
add transformed attention consistency on top of CCL (ablation)TAC-CCL · Stanford Online ProductsRecall@1 (%): 63.0 -> 63.9Improved
add contrastive clustering loss (ablation)TAC-CCL baseline · Cars-196Recall@8 (%): 76.0 -> 75.7Hurt
add contrastive clustering loss (ablation)TAC-CCL baseline · Stanford Online ProductsRecall@100 (%): 87.2 -> 87.2No clear effect
increase number of clusters K beyond 100 (50, 100, 200, 500, 1000) (ablation)TAC-CCL · CUB-200-2011Recall@1 (%)Hurt
increase embedding size from 64 up to 1024 (ablation)TAC-CCL · CUB-200-2011Recall@1 (%)Improved
use TAC-CCL over the best prior unsupervised method (Instance)TAC-CCL · CUB-200-2011Recall@1 (%): 46.2 -> 57.5Improved
use TAC-CCL over the best prior unsupervised method (Instance)TAC-CCL · Cars-196Recall@1 (%): 41.3 -> 46.1Improved

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