Research map / Label-efficient and robust learning / Recognition and long-tail
Graphical Representation for Heterogeneous Face Recognition
Chunlei Peng, Xinbo Gao, Nannan Wang, Jie Li
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 |
|---|---|---|---|
| Fuse 6 similarity metrics from different K values (ablation) | G-HFR (SURF) · CUFSF (viewed sketch) | rank-1 accuracy (%): 92.22 -> 94.24 | Improved |
| Fuse SURF, SIFT and HOG representations (ablation) | G-HFR (SURF) · CUFSF (viewed sketch) | rank-1 accuracy (%): 94.24 -> 96.04 | Improved |
| Fuse SURF, SIFT and HOG representations (ablation) | G-HFR (SIFT) · CUFSF (viewed sketch) | rank-1 accuracy (%): 89.48 -> 96.04 | Improved |
| Fuse SURF, SIFT and HOG representations (ablation) | G-HFR (HOG) · CUFSF (viewed sketch) | rank-1 accuracy (%): 89.05 -> 96.04 | Improved |
| Use Markov network spatial information instead of locally linear embedding (ablation) | G-HFR · CUFSF (viewed sketch) | rank-1 accuracy (%) | Improved |
| Add neighboring compatibility function (ablation) | G-HFR · CUFSF (viewed sketch) | rank-1 accuracy (%) | Improved |
| Use CRSM similarity instead of L-infinity norm (ablation) | G-HFR · CUFSF (viewed sketch) | rank-1 accuracy (%): 1.15% first match rate for L-infinity | Improved |
| Use CRSM similarity instead of L2 or chi-square (ablation) | G-HFR · CUFSF (viewed sketch) | rank-1 accuracy (%) | Improved |
| Vary number of nearest neighbors K from 15 to 80 (ablation) | G-HFR (SURF) · CUFSF (viewed sketch) | rank-1 accuracy (%) | No clear effect |
| G-HFR instead of MCWLD | MCWLD · IIIT-D sketch database | rank-50 accuracy (%): 28.52 -> 30.36 | Improved |
| G-HFR instead of P-RS | P-RS · Forensic sketch database | rank-50 accuracy (%): 20.80 -> 31.96 | Improved |
| G-HFR instead of LRBP | LRBP · CUFSF (viewed sketch) | rank-1 accuracy (%): 91.12 -> 96.04 | Improved |
Closest papers by meaning
- Asymmetric Joint Learning for Heterogeneous Face Recognition (AAAI 2018 · similarity 0.90)
- DVG-Face: Dual Variational Generation for Heterogeneous Face Recognition (TPAMI 2021 · similarity 0.88)
- Dual Variational Generation for Low Shot Heterogeneous Face Recognition (NeurIPS 2019 · similarity 0.87)
- Residual Compensation Networks for Heterogeneous Face Recognition (AAAI 2019 · similarity 0.85)
- Escaping Data Scarcity for High-Resolution Heterogeneous Face Hallucination (CVPR 2022 · similarity 0.84)
- UMFN: Unified Multi-Domain Face Normalization for Joint Cross-domain Prototype Learning and Heterogeneous Face Recognition (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.
