Research map / Image generation / Editing and style transfer
Pseudo Facial Generation With Extreme Poses for Face Recognition
Check 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 |
|---|---|---|---|
| Add pseudo-profile generation to LightCNN-29-v2 | LightCNN-29-v2 · Multi-PIE Setting2, ±90° | Rank-1 (%): 57.82 -> 87.35 | Helped |
| Add pseudo-profile generation to the Multi-PIE-finetuned LightCNN-29-v2 | LightCNN-29-v2* · Multi-PIE Setting2, ±90° | Rank-1 (%): 91.97 -> 93.68 | Helped |
| Add pseudo-profile generation to LightCNN-29-v2 | LightCNN-29-v2 · CFP Frontal-Profile | ACC (%): 92.94 ± 2.00 -> 94.10 ± 2.30 | Helped |
| Add pseudo-profile generation to LightCNN-29-v2 | LightCNN-29-v2 · LFW | ACC (%): +0.02 | Helped |
| Add pseudo-profile generation to ResNet50-SE | ResNet50-SE · IJB-B 1:1 verification, FAR=1e-4 | TAR (%): +0.46 | Helped |
| Add pseudo-profile generation to ResNet50-SE | ResNet50-SE · IJB-C 1:1 verification, FAR=1e-4 | TAR (%): +0.68 | Helped |
| Add pseudo-profile generation to ResNet50-SE | ResNet50-SE · MegaFace Challenge 1, with data refinement | Rank-1 (%): +0.78 | Helped |
| Increase γ from 25 to 1000 (ablation) | PPFG with LightCNN-29-v2 · Multi-PIE Setting2 | Rank-1 AVG (%): 95.73 -> 97.13 | Helped |
| Increase γ from 1000 to 10000 (ablation) | PPFG with LightCNN-29-v2 · Multi-PIE Setting2 | Rank-1 AVG (%): 97.13 -> 97.02 | Hurt |
| Add pseudo-profile generation to LightCNN-29-v2 | LightCNN-29-v2 · Multi-PIE Setting2, ±15° to ±45° | Rank-1 (%): ±15°: 100.00 -> 100.00; ±30°: 100.00 -> 100.00; ±45°: 99.94 -> 99.98 | Mixed |
Closest papers by meaning
- Vec2Face: Scaling Face Dataset Generation with Loosely Constrained Vectors (ICLR 2025 · similarity 0.86)
- HyperFace: Generating Synthetic Face Recognition Datasets by Exploring Face Embedding Hypersphere (ICLR 2025 · similarity 0.85)
- CemiFace: Center-based Semi-hard Synthetic Face Generation for Face Recognition (NeurIPS 2024 · similarity 0.84)
- EFHQ: Multi-purpose ExtremePose-Face-HQ dataset (CVPR 2024 · similarity 0.84)
- HiFi-Portrait: Zero-shot Identity-preserved Portrait Generation with High-fidelity Multi-face Fusion (CVPR 2025 · similarity 0.83)
- GIF: Generative Inspiration for Face Recognition at Scale (CVPR 2025 · similarity 0.83)
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.