AI Paper Atlas

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

Large-Scale Bisample Learning on ID Versus Spot Face Recognition

Xiangyu Zhu, Hao Liu, Zhen Lei, Hailin Shi, Fan Yang, Dong Yi, Guo-Jun Qi, Stan Ziqing Li

IJCV 2019 · Label-efficient and robust learning › Recognition and long-tail · approach: Margin-based face recognition losses · 57 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 verification transfer stage on IvS data after wild pre-training (ablation)CVC · CASIA-IvS-TestVR@FAR=1e-5 (%): 51.90 -> 83.23Improved
add final large-scale DP-softmax classification stage (ablation)CV# · CASIA-IvS-TestVR@FAR=1e-5 (%): 83.23 -> 91.92Improved
add verification stage between the two classification stages (ablation)C#C · CASIA-IvS-TestVR@FAR=1e-5 (%): 72.35 -> 91.92Improved
use sequential verification then classification instead of joint (ablation)C(VC) · CASIA-IvS-TestVR@FAR=1e-5 (%): 82.80 -> 91.92Improved
add wild pre-training before large-scale classification (ablation)##C · CASIA-IvS-TestVR@FAR=1e-5 (%): not converge -> 72.35Improved
use dominant prototypes instead of random prototypes (100k) (ablation)LBL · CASIA-IvS-TestVR@FAR=1e-5 (%): 90.34 -> 91.92Improved
increase random prototypes from 20k to 100k (ablation)RP-softmax · CASIA-IvS-TestVR@FAR=1e-5 (%): 87.84 -> 90.34Improved
increase dominant queue size from 10 to 100 (ablation)DP-softmax · CASIA-IvS-TestVR@FAR=1e-5 (%): 85.12 -> 91.92Improved
increase dominant queue size from 100 to 300 (ablation)DP-softmax · CASIA-IvS-TestVR@FAR=1e-5 (%): 91.92 -> 92.01No clear effect
add dominant queue updating (ablation)DP-softmax · CASIA-IvS-TestVR@FAR=1e-5 (%): 90.77 -> 91.92Improved
use A-softmax instead of softmax (ablation)DP-softmax · CASIA-IvS-TestVR@FAR=1e-5 (%): 89.55 -> 91.92Improved
use ID-prototype instead of avg-prototype with softmax (ablation)random prototype softmax · CASIA-IvS-TestVR@FAR=1e-5 (%): 87.13 -> 89.55Improved

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