Research map / Label-efficient and robust learning / Noisy labels
Co-Mining: Deep Face Recognition With Noisy Labels
Xiaobo Wang, Shuo Wang, Jun Wang, Hailin Shi, Tao Mei
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add noisy labels detection (NLD) (ablation) | Arc-Softmax baseline · CASIA-Clean-R synthetic symmetric noise 0.1 | average accuracy (%): 90.90 -> 91.11 | Improved |
| add high-confidence clean faces exchanging (HCCFE) (ablation) | Arc-Softmax + NLD · CASIA-Clean-R synthetic symmetric noise 0.1 | average accuracy (%): 91.11 -> 91.39 | Improved |
| add clean faces re-weighting (CFR) (ablation) | Arc-Softmax + NLD + HCCFE · CASIA-Clean-R synthetic symmetric noise 0.1 | average accuracy (%): 91.39 -> 92.16 | Improved |
| add noisy labels detection (NLD) (ablation) | Arc-Softmax baseline · CASIA-Clean-R synthetic symmetric noise 0.2 | average accuracy (%): 69.70 -> 77.04 | Improved |
| add high-confidence clean faces exchanging (HCCFE) (ablation) | Arc-Softmax + NLD · CASIA-Clean-R synthetic symmetric noise 0.2 | average accuracy (%): 77.04 -> 78.58 | Improved |
| add clean faces re-weighting (CFR) (ablation) | Arc-Softmax + NLD + HCCFE · CASIA-Clean-R synthetic symmetric noise 0.2 | average accuracy (%): 78.58 -> 83.95 | Improved |
| add noisy labels detection (NLD) (ablation) | Arc-Softmax baseline · CASIA-Clean-R synthetic symmetric noise 0.3 | average accuracy (%): 58.81 -> 67.24 | Improved |
| add high-confidence clean faces exchanging (HCCFE) (ablation) | Arc-Softmax + NLD · CASIA-Clean-R synthetic symmetric noise 0.3 | average accuracy (%): 67.24 -> 72.92 | Improved |
| add clean faces re-weighting (CFR) (ablation) | Arc-Softmax + NLD + HCCFE · CASIA-Clean-R synthetic symmetric noise 0.3 | average accuracy (%): 72.92 -> 81.32 | Improved |
| train on cleaned CASIA instead of noisy original | Arc-Softmax baseline · CASIA (LFW/CALFW/CPLFW/AgeDB/CFP) | average accuracy (%): 91.05 -> 91.73 | Improved |
| apply Co-Mining instead of baseline | MobileFaceNet + Arc-Softmax · Trained on VggFace2-R, CALFW/CPLFW/AgeDB/CFP | average accuracy (%): 91.18 -> 92.07 | Improved |
| apply prior noisy-label methods (MN, DC, CT, CT+) instead of baseline | MobileFaceNet + Arc-Softmax · Trained on VggFace2-R, CALFW/CPLFW/AgeDB/CFP | average accuracy (%): 91.18 -> 90.62 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- Co-Teaching — Co-teaching: Robust training of deep neural networks with extremely noisy labels (NeurIPS 2018) · better on 5 of 5 shared results
- MentorNet — MentorNet: Learning Data-Driven Curriculum for Very Deep Neural Networks on Corrupted Labels (ICML 2018) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Noise-Robust Face Recognition via Non-target Similarity Distribution Guided Sample Selection (ECCV 2026 · similarity 0.89)
- The Devil of Face Recognition is in the Noise (ECCV 2018 · similarity 0.89)
- RepFace: Refining Closed-Set Noise with Progressive Label Correction for Face Recognition (AAAI 2025 · similarity 0.88)
- Noise-Tolerant Paradigm for Training Face Recognition CNNs (CVPR 2019 · similarity 0.88)
- Toward Robustness against Label Noise in Training Deep Discriminative Neural Networks (NeurIPS 2017 · similarity 0.88)
- BoundaryFace: A Mining Framework with Noise Label Self-Correction for Face Recognition (ECCV 2022 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
