Sparsity-Inducing Divergence Losses for Biometric Verification

Open paper
Benchmark Model Rank Results
face-verification-on-ijb-bQ-Margin–TAR@FAR=0.0001: 96.44
face-verification-on-ijb-cQ-Margin–TAR @ FAR=1e-4: 97.78%

Rank counts only results with a code link.