AI Paper Atlas

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

Introspective Deep Metric Learning

Chengkun Wang, Wenzhao Zheng, Zheng Dong Zhu, Jie Zhou, Jiwen Lu

TPAMI 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 41 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 IDML to ProxyAnchor lossProxyAnchor · CUB-200-2011Recall@1: 69.0 -> 70.7Improved
add IDML to ProxyAnchor lossProxyAnchor · Cars196Recall@1: 87.3 -> 90.6Improved
add IDML to ProxyAnchor lossProxyAnchor · Stanford Online ProductsRecall@1: 79.5 -> 81.5Improved
add IDML to triplet loss with semi-hard samplingTriplet-SH · Cars196Recall@1: 70.8 -> 73.7Improved
add Mixup (without ISM) to ProxyAnchor (ablation)ProxyAnchor · CUB-200-2011Recall@1: 69.0 -> 69.8Improved
add ISM (without Mixup) to ProxyAnchor (ablation)ProxyAnchor · Cars196Recall@1: 87.3 -> 88.8Improved
combine ISM with Mixup on top of ISM alone (ablation)ProxyAnchor · Cars196Recall@1: 88.8 -> 90.6Improved
replace IDML by probabilistic embedding learning (PEL) (ablation)ProxyAnchor · CUB-200-2011Recall@1: 69.0 -> 64.9Hurt
use IDML instead of vMF distributional modeling (ablation)ProxyAnchor · CUB-200-2011Recall@1: 70.0 -> 70.7Improved
use ||u1+u2|| instead of ||u1||+||u2|| for uncertainty level (ablation)IDML-PA · CUB-200-2011Recall@1: 70.1 -> 70.7Improved
use similarity-based ISM (ISM-Sim) instead of distance-based (ISM-Dis) (ablation)IDML-PA · Cars196Recall@1: 89.4 -> 90.6Improved
use learned uncertainty embedding instead of random noise vectors (ablation)IDML-PA · CUB-200-2011Recall@1: 70.0 -> 70.7Improved

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