Computational Intelligence and Neuroscience / 2018 / Article / Tab 3

Research Article

A Community Detection Approach to Cleaning Extremely Large Face Database

Table 3

Comparison with other state-of-the-art results on the LFW benchmark (performance in %).

#NetPairwise verificationClose-set identificationOpen-set identification
AccVR@FAR = 0.1%Rank 1DIR@FAR = 1%DIR@FAR = 0.1%

DeepID2+2599.47-95.0080.70-
Face++499.50----
DeepID32599.53-96.0081.40-
IDL Single Model799.6899.1197.6094.1289.08
IDL Ensemble Model7099.7799.4198.0395.8092.09

DeepFace197.35-64.9044.50-
WebFace197.7380.26--28.90
Facebook-WST Fusion198.37-82.5061.90-
CenterLoss198.70-94.0569.97-
Light CNN-29199.33-97.3393.62-
FaceNet199.63----
Ours199.6799.5398.3195.4086.31

We are committed to sharing findings related to COVID-19 as quickly as possible. We will be providing unlimited waivers of publication charges for accepted research articles as well as case reports and case series related to COVID-19. Review articles are excluded from this waiver policy. Sign up here as a reviewer to help fast-track new submissions.