Research map / Label-efficient and robust learning / Recognition and long-tail
Wasserstein CNN: Learning Invariant Features for NIR-VIS Face Recognition
Ran He, Xiang Wu, Zhenan Sun, Tieniu Tan
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 low-rank constraint (ablation) | WCNN · CASIA NIR-VIS 2.0 | Rank-1 accuracy: 98.4 -> 98.7 | Improved |
| add low-rank constraint (ablation) | WCNN · CASIA NIR-VIS 2.0 | VR@FAR=0.1%: 97.6 -> 98.4 | Improved |
| add low-rank constraint (ablation) | WCNN · Oulu-CASIA NIR-VIS | Rank-1 accuracy: 96.4 -> 98.0 | Improved |
| add low-rank constraint (ablation) | WCNN · Oulu-CASIA NIR-VIS | VR@FAR=0.1%: 50.9 -> 54.6 | Improved |
| add low-rank constraint (ablation) | WCNN · BUAA VisNir | Rank-1 accuracy: 95.4 -> 97.4 | Improved |
| add low-rank constraint (ablation) | WCNN · BUAA VisNir | VR@FAR=0.1%: 86.9 -> 91.9 | Improved |
| add low-rank constraint (ablation) | IDR · Oulu-CASIA NIR-VIS | VR@FAR=0.1%: 46.2 -> 50.3 | Improved |
| add low-rank constraint (ablation) | IDR · BUAA VisNir | VR@FAR=0.1%: 84.7 -> 86.0 | Improved |
| replace IDR with WCNN (Wasserstein distance based invariant feature learning) | IDR · CASIA NIR-VIS 2.0 | VR@FAR=0.1%: 95.7 -> 97.6 | Improved |
| replace IDR with WCNN | IDR · Oulu-CASIA NIR-VIS | Rank-1 accuracy: 94.3 -> 96.4 | Improved |
| replace IDR with WCNN | IDR · BUAA VisNir | VR@FAR=0.1%: 84.7 -> 86.9 | Improved |
| train on NIR-VIS data instead of using VIS-only deep models | VGG · CASIA NIR-VIS 2.0 | Rank-1 accuracy: 62.1 -> 98.7 | Improved |
Closest papers by meaning
- Learning Invariant Deep Representation for NIR-VIS Face Recognition (AAAI 2017 · similarity 0.91)
- Adversarial Cross-Spectral Face Completion for NIR-VIS Face Recognition (TPAMI 2019 · similarity 0.87)
- LAMP-HQ: A Large-Scale Multi-pose High-Quality Database and Benchmark for NIR-VIS Face Recognition (IJCV 2021 · similarity 0.86)
- Residual Compensation Networks for Heterogeneous Face Recognition (AAAI 2019 · similarity 0.86)
- Disentangled Variational Representation for Heterogeneous Face Recognition (AAAI 2019 · similarity 0.86)
- DVG-Face: Dual Variational Generation for Heterogeneous Face Recognition (TPAMI 2021 · similarity 0.86)
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.
