Harnessing Geometric Constraints from Emotion Labels to improve Face Verification
This work addresses face verification accuracy by leveraging emotion labels, but it is incremental as it builds on existing loss functions without altering the neural network backbone.
The paper tackles face verification by using auxiliary facial emotion labels to impose geometric constraints on the embedding space, resulting in improved performance with novel loss functions that integrate with standard methods like Triplet Loss or ArcFace.
For the task of face verification, we explore the utility of harnessing auxiliary facial emotion labels to impose explicit geometric constraints on the embedding space when training deep embedding models. We introduce several novel loss functions that, in conjunction with a standard Triplet Loss [43], or ArcFace loss [10], provide geometric constraints on the embedding space; the labels for our loss functions can be provided using either manually annotated or automatically detected auxiliary emotion labels. Our method is implemented purely in terms of the loss function and does not require any changes to the neural network backbone of the embedding function.