GhostVLAD for set-based face recognition

23 Oct 2018Yujie ZhongRelja ArandjelovićAndrew Zisserman

The objective of this paper is to learn a compact representation of image sets for template-based face recognition. We make the following contributions: first, we propose a network architecture which aggregates and embeds the face descriptors produced by deep convolutional neural networks into a compact fixed-length representation... (read more)

PDF Abstract

Evaluation results from the paper

Task Dataset Model Metric name Metric value Global rank Compare
Face Recognition IJB-A SE-GV-3 TAR @ FAR=0.01 0.972 # 1
Face Verification IJB-A SE-GV-4-g1 TAR @ FAR=0.01 97.2 # 2
Face Recognition IJB-B GhostVLAD, SE-GV-3 TAR @ FAR=0.01 0.963 # 1
Face Verification IJB-B SE-GV-3-g2 TAR @ FAR=0.01 96.4 # 2