AI Paper Atlas

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

DAM: Discrepancy Alignment Metric for Face Recognition

Jiaheng Liu, Yudong Wu, Yichao Wu, Chuming Li, Xiaolin Hu, Ding Liang, Mengyu Wang

ICCV 2021 · Label-efficient and robust learning › Recognition and long-tail · approach: Margin-based face recognition losses · 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 DAM-R verification metricCosFace · IJB-B (TAR@FAR=1e-4)TAR (%): 94.80 -> 94.97Improved
add DAM-R verification metricArcFace · IJB-B (TAR@FAR=1e-4)TAR (%): 94.25 -> 94.63Improved
add DAM-R verification metricCurricularFace · IJB-C (TAR@FAR=1e-4)TAR (%): 96.11 -> 96.20Improved
add DAM-R verification metricArcFace ResNet-18 · LFWverification accuracy (%): 98.73 -> 99.03Improved
add DAM-R verification metricArcFace ResNet-18 · CALFWverification accuracy (%): 91.67 -> 92.05Improved
add DAM-R verification metricArcFace ResNet-18 · YTFverification accuracy (%): 94.97 -> 95.23Improved
add DAM-R verification metricArcFace ResNet-34 · RFW Asianverification accuracy (%): 93.75 -> 94.31Improved
add DAM-R verification metricArcFace ResNet-34 · RFW averageverification accuracy (%): 94.63 -> 95.08Improved
use learned LID estimator (DAM-L) instead of reference-based (DAM-R) (ablation)DAM · IJB-B (TAR@FAR=1e-4)TAR (%): 94.63 -> 94.54No clear effect
use real-image anchors instead of FC-layer weights (ablation)DAM-R · IJB-B (TAR@FAR=0.001%)TAR (%): 87.50 -> 87.86No clear effect
use StyleGAN anchors instead of real images (ablation)DAM-R · IJB-B (TAR@FAR=0.01%)TAR (%): 93.64 -> 93.63No clear effect
replace summation with multiplication in combining LID (ablation)DAM-R · IJB-B (TAR@FAR=1e-4)TAR (%): 94.63 -> 94.56No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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