Dynamic Training Data Dropout for Robust Deep Face Recognition | IEEE Journals & Magazine | IEEE Xplore

Dynamic Training Data Dropout for Robust Deep Face Recognition


Abstract:

Learning with noise is a practically challenging problem in deep face recognition. Despite the success of large margin softmax loss functions, these methods are designed ...Show More

Abstract:

Learning with noise is a practically challenging problem in deep face recognition. Despite the success of large margin softmax loss functions, these methods are designed for clean face databases. Considering the inevitable noise in the large scale databases, we first analyze the performance of noise in the training databases. For noise-robust deep face recognition, we propose a dynamic training data dropout (DTDD) method to dynamically filter the noise in the training database and gradually form a stable refined database for model learning. Specifically, we leverage the information provided by the model predictions of accumulated training epochs, which can distinguish regular samples and noise effectively and accurately. The proposed DTDD method is easy and stable for implementation, and can be combined with existing state-of-the-art loss functions and network architectures. Extensive experiments on CASIA-WebFace, VGGFace2, and MS-Celeb-1 M databases empirically demonstrate that our proposed method can robustly train deep face recognition models in the presence of label noise and low quality images.
Published in: IEEE Transactions on Multimedia ( Volume: 24)
Page(s): 1186 - 1197
Date of Publication: 27 October 2021

ISSN Information:

Funding Agency:


I. Introduction

Deep learning techniques have achieved considerable success in computer vision [1]–[5], significantly improving state-of-the-art in face recognition [6]. In addition to the evolving architectures [3], [7]–[11] and effective loss functions [8], [12]–[18], large-scale training databases [11], [19]–[21] play a crucial role in deep face recognition.

References

References is not available for this document.