Global-Local GCN: Large-Scale Label Noise Cleansing for Face Recognition

By Yaobin Zhang, Weihong Deng, Mei Wang, Jiani Hu, Xian Li, Dongyue Zhao, Dongchao Wen. 2020

74 citations recorded by the source corpus.

In the field of face recognition, large-scale web-collected datasets are essential for learning discriminative representations, but they suffer from noisy identity labels, such as outliers and label flips. It is beneficial to automatically cleanse their label noise for improving recognition accuracy. Unfortunately, existing cleansing methods cannot accurately identify noise in the wild. To solve this problem, we propose an effective automatic label noise cleansing framework for face recognition datasets, FaceGraph. Using two cascaded graph convolutional networks, FaceGraph performs global-to-local discrimination to select useful data in a noisy environment. Extensive experiments show that cleansing widely used datasets, such as CASIA-WebFace, VGGFace2, MegaFace2, and MS-Celeb-1M, using the proposed method can improve the recognition performance of state-of-the-art representation learning methods like Arcface. Further, we cleanse massive self-collected celebrity data, namely MillionCelebs, to provide 18.8M images of 636K identities. Training with the new data, Arcface surpasses state-of-the-art performance by a notable margin to reach 95.62% TPR at 1e-5 FPR on the IJB-C benchmark.

Subfields: Computer Vision and Pattern Recognition, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Jie Zhou
2. Jiwen Lu
3. Guan Huang
4. Junjie Huang
5. Xiaogang Wang
6. Dalong Du
7. Luc Van Gool
8. Zheng Zhu

In Signal Processing

1. Jie Zhou
2. Jiwen Lu
3. Gunhee Kim
4. Jiankang Deng
5. Chris Dongjoo Kim
6. Dalong Du
7. Stefanos P. Zafeiriou
8. Daniel Cohen-Or

Citations in the corpus

• Continual Learning on Noisy Data Streams via Self-Purified Replay (2021, background): “Some works use graphs for offline learning from a large-scale noisy dataset [95, 93].”
• WebFace260M: A Benchmark Unveiling the Power of Million-Scale Deep Face Recognition (2021, methodology): “Compared with the most recent GCNbased cleaning [87], the data cleaned by the CAST also achieves higher performance.”
• Deep Face Recognition: A Survey (2018, unknown): “8/804 celebrity MillionCelebs [165] 2020 18.”
• DualGraph: A graph-based method for reasoning about label noise (2021, methodology): “FaceGraph [39] uses two cascaded GCNs to select useful data in the large-scale web-collected face datasets.”