Angular Learning: Toward Discriminative Embedded Features
This work addresses overfitting and hyperparameter tuning issues in feature embedding for face recognition and object classification, presenting an incremental improvement with a simpler control mechanism.
The paper tackled the problem of overfitting and hyperparameter sensitivity in margin-based softmax loss functions for face recognition and object classification by proposing an angular loss method that maximizes angular gradient to promote intra-class compactness, resulting in improved accuracy, discriminative information, and reduced time-consumption on well-known datasets.
The margin-based softmax loss functions greatly enhance intra-class compactness and perform well on the tasks of face recognition and object classification. Outperformance, however, depends on the careful hyperparameter selection. Moreover, the hard angle restriction also increases the risk of overfitting. In this paper, angular loss suggested by maximizing the angular gradient to promote intra-class compactness avoids overfitting. Besides, our method has only one adjustable constant for intra-class compactness control. We define three metrics to measure inter-class separability and intra-class compactness. In experiments, we test our method, as well as other methods, on many well-known datasets. Experimental results reveal that our method has the superiority of accuracy improvement, discriminative information, and time-consumption.