Research map / Label-efficient and robust learning / Recognition and long-tail
Towards Pose Invariant Face Recognition in the Wild
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Jointly frontalize faces and learn discriminative representations with PIM | Light CNN-29 · Multi-PIE Setting-1, ±90° | rank-1 recognition rate (%): 64.03 -> 75.00 | Helped |
| Jointly frontalize faces and learn discriminative representations with PIM | Light CNN-29 · Multi-PIE Setting-2, ±90° | rank-1 recognition rate (%): 64.64 -> 86.50 | Helped |
| Use PIM rather than the Light CNN-29 baseline | CFP Frontal-Profile · CFP Frontal-Profile | EER (%): 8.71±1.80 -> 7.69±1.29 | Helped |
| Use PIM rather than Light CNN-29 | CFP Frontal-Frontal · CFP Frontal-Frontal | accuracy (%): 99.64±0.32 -> 99.44±0.36 | No clear effect |
| Add pixel-wise content loss (ablation) | PIM · Multi-PIE Setting-1, ±90° | rank-1 recognition rate (%): 60.60 -> 75.00 | Helped |
| Add the local-path generator (ablation) | PIM · Multi-PIE Setting-1, ±90° | rank-1 recognition rate (%): 66.80 -> 75.00 | Helped |
| Add the siamese discriminator learner branch (ablation) | PIM · Multi-PIE Setting-1, ±90° | rank-1 recognition rate (%): 66.90 -> 75.00 | Helped |
| Add dynamic convolution (ablation) | PIM · Multi-PIE Setting-1, ±90° | rank-1 recognition rate (%): 69.80 -> 75.00 | Helped |
| Add cross-domain adversarial training (ablation) | PIM · Multi-PIE Setting-1, ±90° | rank-1 recognition rate (%): 71.10 -> 75.00 | Helped |
| Add symmetry loss (ablation) | PIM · Multi-PIE Setting-1, ±90° | rank-1 recognition rate (%): 72.30 -> 75.00 | Helped |
Closest papers by meaning
- Adversarial Pose Regression Network for Pose-Invariant Face Recognitions (AAAI 2021 · similarity 0.91)
- Pose-Aware Face Recognition in the Wild (CVPR 2016 · similarity 0.89)
- Pose-Invariant Face Recognition via Adaptive Angular Distillation (AAAI 2022 · similarity 0.89)
- Disentangled Representation Learning GAN for Pose-Invariant Face Recognition (CVPR 2017 · similarity 0.89)
- Discriminative Invariant Kernel Features: A Bells-and-Whistles-Free Approach to Unsupervised Face Recognition and Pose Estimation (CVPR 2016 · similarity 0.88)
- Dual Pose-invariant Embeddings: Learning Category and Object-specific Discriminative Representations for Recognition and Retrieval (CVPR 2024 · 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.