# Noise-Tolerant Paradigm for Training Face Recognition CNNs

arXiv: https://arxiv.org/abs/1903.10357 · HTML: https://mlsanity.com/paper/1903.10357/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Mar 2019.

## Built on by (most cited first, 20 of 20)

- [Dive into Ambiguity: Latent Distribution Mining and Pairwise Uncertainty Estimation for Facial Expression Recognition](https://mlsanity.com/paper/2104.00232/) (2021)
- [UNICON: Combating Label Noise Through Uniform Selection and Contrastive Learning](https://mlsanity.com/paper/2203.14542/) (2022)
- [Killing Two Birds with One Stone:Efficient and Robust Training of Face Recognition CNNs by Partial FC](https://mlsanity.com/paper/2203.15565/) (2022)
- [WebFace260M: A Benchmark for Million-Scale Deep Face Recognition](https://mlsanity.com/paper/2204.10149/) (2022)
- [Simple and Robust Loss Design for Multi-Label Learning with Missing Labels](https://mlsanity.com/paper/2112.07368/) (2021)
- [SFace: Sigmoid-Constrained Hypersphere Loss for Robust Face Recognition](https://mlsanity.com/paper/2205.12010/) (2022)
- [EQFace: A Simple Explicit Quality Network for Face Recognition](https://mlsanity.com/paper/2105.00634/) (2021)
- [Uncertain Facial Expression Recognition via Multi-task Assisted Correction](https://mlsanity.com/paper/2212.07144/) (2022)
- [Uncertain Label Correction via Auxiliary Action Unit Graphs for Facial Expression Recognition](https://mlsanity.com/paper/2204.11053/) (2022)
- [DAIL: Dataset-Aware and Invariant Learning for Face Recognition](https://mlsanity.com/paper/2101.05419/) (2021)
- [Do We Really Need Gold Samples for Sample Weighting Under Label Noise?](https://mlsanity.com/paper/2104.09045/) (2021)
- [Median Pixel Difference Convolutional Network for Robust Face Recognition](https://mlsanity.com/paper/2205.15867/) (2022)
- [Towards Robust Face Recognition with Comprehensive Search](https://mlsanity.com/paper/2208.13600/) (2022)
- [QCFace: Image Quality Control for boosting Face Representation & Recognition](https://mlsanity.com/paper/2510.15289/) (2025)
- [Basket-based Softmax](https://mlsanity.com/paper/2201.09308/) (2022)
- [Rethinking Robust Representation Learning Under Fine-grained Noisy Faces](https://mlsanity.com/paper/2208.04352/) (2022)
- [A Survey of Face Recognition](https://mlsanity.com/paper/2212.13038/) (2022)
- [RepFace: Refining Closed-Set Noise with Progressive Label Correction for Face Recognition](https://mlsanity.com/paper/2412.12031/) (2024)
- [Enhancing Sample Utilization in Noise-Robust Deep Metric Learning With Subgroup-Based Positive-Pair Selection](https://mlsanity.com/paper/2501.11063/) (2025)
- [FaceFusion: Exploiting Full Spectrum of Multiple Datasets](https://mlsanity.com/paper/2305.14601/) (2023)

## Abstract (arXiv)

Benefit from large-scale training datasets, deep Convolutional Neural Networks(CNNs) have achieved impressive results in face recognition(FR). However, tremendous scale of datasets inevitably lead to noisy data, which obviously reduce the performance of the trained CNN models. Kicking out wrong labels from large-scale FR datasets is still very expensive, although some cleaning approaches are proposed. According to the analysis of the whole process of training CNN models supervised by angular margin based loss(AM-Loss) functions, we find that the θ distribution of training samples implicitly reflects their probability of being clean. Thus, we propose a novel training paradigm that employs the idea of weighting samples based on the above probability. Without any prior knowledge of noise, we can train high performance CNN models with large-scale FR datasets. Experiments demonstrate the effectiveness of our training paradigm. The codes are available at https://github.com/huangyangyu/NoiseFace.