Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

MagFace: A Universal Representation for Face Recognition and Quality Assessment

About

The performance of face recognition system degrades when the variability of the acquired faces increases. Prior work alleviates this issue by either monitoring the face quality in pre-processing or predicting the data uncertainty along with the face feature. This paper proposes MagFace, a category of losses that learn a universal feature embedding whose magnitude can measure the quality of the given face. Under the new loss, it can be proven that the magnitude of the feature embedding monotonically increases if the subject is more likely to be recognized. In addition, MagFace introduces an adaptive mechanism to learn a wellstructured within-class feature distributions by pulling easy samples to class centers while pushing hard samples away. This prevents models from overfitting on noisy low-quality samples and improves face recognition in the wild. Extensive experiments conducted on face recognition, quality assessments as well as clustering demonstrate its superiority over state-of-the-arts. The code is available at https://github.com/IrvingMeng/MagFace.

Qiang Meng, Shichao Zhao, Zhida Huang, Feng Zhou• 2021

Related benchmarks

TaskDatasetResultRank
Face VerificationLFW
Mean Accuracy99.83
417
Face VerificationAgeDB-30
Accuracy98.17
256
Face VerificationCPLFW
Accuracy92.87
255
Face VerificationIJB-C
TAR @ FAR=0.01%95.97
243
Face RecognitionLFW
Accuracy99.73
229
Face VerificationCALFW
Accuracy96.15
209
Face VerificationCFP-FP
Accuracy98.46
206
Face VerificationIJB-B
TAR (FAR=1e-4)94.51
201
Face VerificationLFW (test)
Verification Accuracy99.83
173
Face RecognitionCFP-FP
Accuracy98.46
121
Showing 10 of 56 rows

Other info

Code

Follow for update