MixFace: Improving Face Verification Focusing on Fine-grained Conditions
This work addresses the challenge of improving face verification for fine-grained conditions, representing an incremental advance in the field.
The authors tackled the problem of face recognition under fine-grained conditions by proposing MixFace, a novel loss function combining classification and metric losses, which demonstrated effectiveness and robustness across various benchmark datasets.
The performance of face recognition has become saturated for public benchmark datasets such as LFW, CFP-FP, and AgeDB, owing to the rapid advances in CNNs. However, the effects of faces with various fine-grained conditions on FR models have not been investigated because of the absence of such datasets. This paper analyzes their effects in terms of different conditions and loss functions using K-FACE, a recently introduced FR dataset with fine-grained conditions. We propose a novel loss function, MixFace, that combines classification and metric losses. The superiority of MixFace in terms of effectiveness and robustness is demonstrated experimentally on various benchmark datasets.