Research map / Label-efficient and robust learning / Recognition and long-tail
When Age-Invariant Face Recognition Meets Face Age Synthesis: A Multi-Task Learning Framework and a New Benchmark
Zhizhong Huang, Junping Zhang, Hongming Shan
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 age estimation loss (ablation) | CosFace baseline · AgeDB-30 | accuracy (%): 95.52 -> 95.32 | No clear effect |
| add attention-based feature decomposition (AFD) (ablation) | CosFace baseline · AgeDB-30 | accuracy (%): 95.52 -> 95.90 | Improved |
| add AFD (ablation) | CosFace baseline · FG-NET | rank-1 (%): 93.64 -> 94.58 | Improved |
| parallel channel+spatial attention in AFD instead of CBAM (ablation) | AFD · AgeDB-30 | accuracy (%): 96.08 -> 95.90 | Mixed |
| add cross-age domain adversarial training (ablation) | MTLFace (AFD) · AgeDB-30 | accuracy (%): 95.90 -> 96.23 | Improved |
| fine-tune on all synthesized child faces (FT-All) (ablation) | MTLFace w/o FT · AgeDB-30 | accuracy (%): 96.23 -> 95.88 | Hurt |
| fine-tune on selected synthesized faces (FT-Sel) (ablation) | MTLFace w/o FT · AgeDB-30 | accuracy (%): 96.23 -> 96.45 | Improved |
| fine-tune on selected synthesized faces (FT-Sel) (ablation) | MTLFace w/o FT · FG-NET | rank-1 (%): 94.78 -> 95.00 | Improved |
| add AIFR task to face age synthesis (ablation) | MTLFace · MORPH | age accuracy (%): 51.79 -> 67.07 | Improved |
| add AIFR task to face age synthesis (ablation) | MTLFace · MORPH | identity cosine similarity: 0.219 -> 0.652 | Improved |
| add multi-level ICMs (ablation) | MTLFace · MORPH | age accuracy (%): 64.80 -> 67.07 | Improved |
| add ICM (ablation) | MTLFace · ECAF | MAE: 3.16 -> 2.28 | Improved |
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.
- ArcFace — ArcFace: Additive Angular Margin Loss for Deep Face Recognition (CVPR 2019) · better on 4 of 4 shared results
- CosFace — CosFace: Large Margin Cosine Loss for Deep Face Recognition (CVPR 2018) · better on 3 of 3 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
- When Age-Invariant Face Recognition Meets Face Age Synthesis: A Multi-Task Learning Framework (CVPR 2021 · similarity 0.98)
- Towards Age-Invariant Face Recognition (TPAMI 2020 · similarity 0.94)
- Look across Elapse: Disentangled Representation Learning and Photorealistic Cross-Age Face Synthesis for Age-Invariant Face Recognition (AAAI 2019 · similarity 0.92)
- Orthogonal Deep Features Decomposition for Age-Invariant Face Recognition (ECCV 2018 · similarity 0.91)
- From Cradle to Cane: A Two-Pass Framework for High-Fidelity Lifespan Face Aging (NeurIPS 2025 · similarity 0.89)
- AQUAFace: Age-Invariant Quality Adaptive Face Recognition for Unconstrained Selfie vs ID Verification (AAAI 2025 · 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.
