Journal of Terahertz Science and Electronic Information Technology , Volume. 21, Issue 9, 1150(2023)

Partial Area Under Curve optimization for face recognition system

TANGLinruize, BAIZhongxin, and ZHANG Xiaolei
Author Affiliations
  • [in Chinese]
  • show less

    Deep learning based face recognition has outperformed traditional methods in many application scenarios. There are two main lines of research to design loss functions for face recognition, i. e., verification and identification. The verification loss functions match the pipeline of open-set face recognition, but it is hard to implement. Therefore, most state-of-the-art deep learning methods for face recognition take the identification loss functions with softmax output units and cross-entropy loss. Nevertheless, identification loss function dose not match the training process with evaluation procedure. A verification loss function is proposed for open-set face recognition to maximize partial area under the Receiver-Operating-Characteristic(ROC) curve, partial Area Under Curve(pAUC). A class-center learning method is also proposed to improve training efficiency, which is critical for the proposed loss function to be comparable to the identification loss in performance. Experimental results on five large scale unconstrained face recognition benchmarks show that the proposed method is highly competitive with state-of-the-art face recognition methods.

    Tools

    Get Citation

    Copy Citation Text

    TANGLinruize, BAIZhongxin, ZHANG Xiaolei. Partial Area Under Curve optimization for face recognition system[J]. Journal of Terahertz Science and Electronic Information Technology , 2023, 21(9): 1150

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Received: Jun. 18, 2021

    Accepted: --

    Published Online: Jan. 19, 2024

    The Author Email:

    DOI:10.11805/tkyda2021258

    Topics