# The Devil of Face Recognition is in the Noise

arXiv: https://arxiv.org/abs/1807.11649 · HTML: https://mlsanity.com/paper/1807.11649/

## Key facts

- Cited by 65 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Jul 2018.

## Built on by (most cited first, 30 of 65)

- [WebFace260M: A Benchmark Unveiling the Power of Million-Scale Deep Face Recognition](https://mlsanity.com/paper/2103.04098/) (2021)
- [SynFace: Face Recognition with Synthetic Data](https://mlsanity.com/paper/2108.07960/) (2021)
- [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)
- [Synthetic Data for Face Recognition: Current State and Future Prospects](https://mlsanity.com/paper/2305.01021/) (2023)
- [Pre-training strategies and datasets for facial representation learning](https://mlsanity.com/paper/2103.16554/) (2021)
- [Federated Learning for Face Recognition with Gradient Correction](https://mlsanity.com/paper/2112.07246/) (2021)
- [Compatibility-aware Heterogeneous Visual Search](https://mlsanity.com/paper/2105.06047/) (2021)
- [SFace: Sigmoid-Constrained Hypersphere Loss for Robust Face Recognition](https://mlsanity.com/paper/2205.12010/) (2022)
- [Deep Learning for Camera Calibration and Beyond: A Survey](https://mlsanity.com/paper/2303.10559/) (2023)
- [Joint Face Image Restoration and Frontalization for Recognition](https://mlsanity.com/paper/2105.09907/) (2021)
- [Learning Hierarchical Graph Neural Networks for Image Clustering](https://mlsanity.com/paper/2107.01319/) (2021)
- [Neuromorphic Event-based Facial Expression Recognition](https://mlsanity.com/paper/2304.06351/) (2023)
- [Cross-Domain Similarity Learning for Face Recognition in Unseen Domains](https://mlsanity.com/paper/2103.07503/) (2021)
- [The METRIC-framework for assessing data quality for trustworthy AI in medicine: a systematic review](https://mlsanity.com/paper/2402.13635/) (2024)
- [Fairness and Bias Mitigation in Computer Vision: A Survey](https://mlsanity.com/paper/2408.02464/) (2024)
- [An Efficient Training Approach for Very Large Scale Face Recognition](https://mlsanity.com/paper/2105.10375/) (2021)
- [Masked Face Recognition Challenge: The WebFace260M Track Report](https://mlsanity.com/paper/2108.07189/) (2021)
- [Exploring Robust Architectures for Deep Artificial Neural Networks](https://mlsanity.com/paper/2106.15850/) (2021)
- [Synthesizing Efficient Data with Diffusion Models for Person Re-Identification Pre-Training](https://mlsanity.com/paper/2406.06045/) (2024)
- [EQFace: A Simple Explicit Quality Network for Face Recognition](https://mlsanity.com/paper/2105.00634/) (2021)
- [Towards Robustness to Label Noise in Text Classification via Noise Modeling](https://mlsanity.com/paper/2101.11214/) (2021)
- [Dual-Tuning: Joint Prototype Transfer and Structure Regularization for Compatible Feature Learning](https://mlsanity.com/paper/2108.02959/) (2021)
- [Beyond Real Faces: Synthetic Datasets Can Achieve Reliable Recognition Performance without Privacy Compromise](https://mlsanity.com/paper/2510.17372/) (2025)
- [Harnessing Unrecognizable Faces for Improving Face Recognition](https://mlsanity.com/paper/2106.04112/) (2021)
- [BoundaryFace: A mining framework with noise label self-correction for Face Recognition](https://mlsanity.com/paper/2210.04567/) (2022)
- [Metric Compatible Training for Online Backfilling in Large-Scale Retrieval](https://mlsanity.com/paper/2301.03767/) (2023)
- [Recent Advances in Deep Learning Techniques for Face Recognition](https://mlsanity.com/paper/2103.10492/) (2021)
- [VIGFace: Virtual Identity Generation for Privacy-Free Face Recognition](https://mlsanity.com/paper/2403.08277/) (2024)
- [HiFi-Portrait: Zero-shot Identity-preserved Portrait Generation with High-fidelity Multi-face Fusion](https://mlsanity.com/paper/2512.14542/) (2025)

## Abstract (arXiv)

The growing scale of face recognition datasets empowers us to train strong convolutional networks for face recognition. While a variety of architectures and loss functions have been devised, we still have a limited understanding of the source and consequence of label noise inherent in existing datasets. We make the following contributions: 1) We contribute cleaned subsets of popular face databases, i.e., MegaFace and MS-Celeb-1M datasets, and build a new large-scale noise-controlled IMDb-Face dataset. 2) With the original datasets and cleaned subsets, we profile and analyze label noise properties of MegaFace and MS-Celeb-1M. We show that a few orders more samples are needed to achieve the same accuracy yielded by a clean subset. 3) We study the association between different types of noise, i.e., label flips and outliers, with the accuracy of face recognition models. 4) We investigate ways to improve data cleanliness, including a comprehensive user study on the influence of data labeling strategies to annotation accuracy. The IMDb-Face dataset has been released on https://github.com/fwang91/IMDb-Face.