RepFace: Refining Closed-Set Noise with Progressive Label Correction for Face Recognition

TL;DR

This paper introduces RepFace, a novel framework for face recognition that effectively handles closed-set label noise through progressive label correction, sample grouping, and label smoothing, leading to state-of-the-art results.

Contribution

The paper proposes a new training strategy that stabilizes early-stage training and refines labels for noisy data, improving robustness against closed-set label noise in face recognition.

Findings

Achieves state-of-the-art accuracy on mainstream face datasets.

Effectively distinguishes clean, ambiguous, and noisy samples during training.

Demonstrates robustness to closed-set label noise with improved training stability.

Abstract

Face recognition has made remarkable strides, driven by the expanding scale of datasets, advancements in various backbone and discriminative losses. However, face recognition performance is heavily affected by the label noise, especially closed-set noise. While numerous studies have focused on handling label noise, addressing closed-set noise still poses challenges. This paper identifies this challenge as training isn't robust to noise at the early-stage training, and necessitating an appropriate learning strategy for samples with low confidence, which are often misclassified as closed-set noise in later training phases. To address these issues, we propose a new framework to stabilize the training at early stages and split the samples into clean, ambiguous and noisy groups which are devised with separate training strategies. Initially, we employ generated auxiliary closed-set noisy samples to enable the model to identify noisy data at the early stages of training. Subsequently, we introduce how samples are split into clean, ambiguous and noisy groups by their similarity to the positive and nearest negative centers. Then we perform label fusion for ambiguous samples by incorporating accumulated model predictions. Finally, we apply label smoothing within the closed set, adjusting the label to a point between the nearest negative class and the initially assigned label. Extensive experiments validate the effectiveness of our method on mainstream face datasets, achieving state-of-the-art results. The code will be released upon acceptance.