MassFace: an efficient implementation using triplet loss for face recognition
This work provides practical insights for researchers and practitioners applying triplet loss in face recognition, though it is incremental as it focuses on optimizing an existing method.
The paper tackled the challenge of efficiently training face recognition models using triplet loss by analyzing factors that influence training, achieving competitive results on the LFW benchmark with models trained on the CASIA-Webface dataset.
In this paper we present an efficient implementation using triplet loss for face recognition. We conduct the practical experiment to analyze the factors that influence the training of triplet loss. All models are trained on CASIA-Webface dataset and tested on LFW. We analyze the experiment results and give some insights to help others balance the factors when they apply triplet loss to their own problem especially for face recognition task. Code has been released in https://github.com/yule-li/MassFace.