ARoFace: Alignment Robustness to Improve Low-Quality Face Recognition
Open paper
Benchmark
Model
Rank
Results
face-verification-on-ijb-b
AdaFace+ARoFace (WebFace4M)
#19
TAR@FAR=0.0001: 96.35
face-verification-on-ijb-c
AdaFace+ARoFace (WebFace12M)
#33
TAR @ FAR=1e-4: 97.6%
Rank-1: 97.61
Rank-5: 98.57
Rank counts only results with a code link.