AI Paper Atlas

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

Permutation-Invariant Feature Restructuring for Correlation-Aware Image Set-Based Recognition

Xiaofeng Liu, Zhenhua Guo, Site Li, Lingsheng Kong, Ping Jia, Jane You, B.V.K. Vijaya Kumar

ICCV 2019 · Label-efficient and robust learning › Recognition and long-tail · approach: Margin-based face recognition losses · 36 citations (OpenAlex)

Read the paper

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 DFA inter-set module (l2) to RSA (ablation)RSA (GoogleNet) · IJB-A 1:1 verification (GoogleNet)TAR at FAR=0.001: 0.926 -> 0.937Improved
add DFA inter-set module (l2) to RSA (ablation)RSA (ResNet50) · IJB-A 1:1 verification (ResNet50)TAR at FAR=0.001: 0.943 -> 0.955Improved
add RSA inner-set module to DFA(l2) (ablation)DFA(l2) (GoogleNet) · IJB-A 1:1 verification (GoogleNet)TAR at FAR=0.001: 0.914 -> 0.937Improved
use l2 collaborative instead of l1 sparse coding in DFA (ablation)PIFR (GoogleNet) · IJB-A 1:1 verification (GoogleNet)TAR at FAR=0.001: 0.933 -> 0.937Improved
add symmetric dictionary learning (x2) (ablation)PIFR(l2) (GoogleNet) · IJB-A 1:1 verification (GoogleNet)TAR at FAR=0.001: 0.937 -> 0.938No clear effect
add DFA to RSA (ablation)RSA (GoogleNet) · Celebrity-1000 open-set identification, 800 subjectsrank-1 accuracy: 0.8558 -> 0.8604Improved
add Gaussian spatial similarity Delta to RSA (ablation)RSA (GoogleNet) · Celebrity-1000 open-set identification, 800 subjectsrank-1 accuracy: 0.8420 -> 0.8558Improved
use feature difference residual instead of neighbor feature as residue (ablation)RSA (GoogleNet) · Celebrity-1000 open-set identification, 800 subjectsrank-1 accuracy: 0.8388 -> 0.8558Improved
replace original non-local with RSA (ablation)non-local (GoogleNet) · Celebrity-1000 open-set identification, 800 subjectsrank-1 accuracy: 0.8251 -> 0.8558Improved
use embedded Gaussian affinity in every RSA block (ablation)RSA (GoogleNet) · Celebrity-1000 open-set identification, 800 subjectsrank-1 accuracy: 0.8556 -> 0.8558No clear effect
switch backbone from GoogleNet to ResNet50PIFR(l2) · Celebrity-1000 open-set identification, 800 subjectsrank-1 accuracy: 0.8604 -> 0.8713Improved
replace DAC with RSADAC (GoogleNet) · IJB-A 1:N identification (GoogleNet)Rank-1 accuracy: 0.973 -> 0.984Improved

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