Activation Template Matching Loss for Explainable Face Recognition
This addresses the need for interpretable AI in face recognition, offering a method to automatically learn facial parts, though it appears incremental as it builds on existing channel-based explainability approaches.
The paper tackled the problem of constructing an explainable face recognition network that learns facial part-based features without manual annotation or additional datasets, proposing an Explainable Channel Loss (ECLoss) that achieves superior explainability metrics and improves face verification performance without alignment.
Can we construct an explainable face recognition network able to learn a facial part-based feature like eyes, nose, mouth and so forth, without any manual annotation or additionalsion datasets? In this paper, we propose a generic Explainable Channel Loss (ECLoss) to construct an explainable face recognition network. The explainable network trained with ECLoss can easily learn the facial part-based representation on the target convolutional layer, where an individual channel can detect a certain face part. Our experiments on dozens of datasets show that ECLoss achieves superior explainability metrics, and at the same time improves the performance of face verification without face alignment. In addition, our visualization results also illustrate the effectiveness of the proposed ECLoss.