Research map / Image generation / Editing and style transfer
AgeTransGAN for Facial Age Transformation with Rectified Performance Metrics
Gee-Sern Hsu, Rui-Cang Xie, Zhi-Ting Chen, Yu-Hong Lin
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 cycle-consistency and triplet losses (ablation) | AgeTransGAN baseline · FFHQ-Aging 20-29 to 0-2 | verification rate: 63.1 -> 77.4 | Improved |
| add triplet loss on top of cycle loss (ablation) | AgeTransGAN · FFHQ-Aging 20-29 to 0-2 | verification rate: 74.37 -> 77.4 | Improved |
| add identity loss to cycle and triplet losses (ablation) | AgeTransGAN · FFHQ-Aging 20-29 to 0-2 | verification rate: 77.4 -> 80.3 | Improved |
| replace Lid with pretrained ArcFace identity loss (ablation) | AgeTransGAN · FFHQ-Aging 20-29 to 0-2 | age mean error (years): 0.4 -> 7.4 | Hurt |
| replace Lid with pretrained ArcFace identity loss (ablation) | AgeTransGAN · FFHQ-Aging 20-29 to 0-2 | verification rate: 80.3 -> 98.3 | Improved |
| add label projection to discriminator (D0+Ca to Dp+Ca) (ablation) | AgeTransGAN · FFHQ-Aging 20-29 to 0-2 | verification rate: 74.7 -> 80.3 | Improved |
| add age classifier Ca to discriminator Dp (ablation) | AgeTransGAN · FFHQ-Aging 20-29 to 0-2 | verification rate: 75.2 -> 80.3 | Improved |
| use multilayer instead of single-layer features in Dp (ablation) | AgeTransGAN · FFHQ-Aging 20-29 to 0-2 | verification rate: 77.7 -> 80.3 | Improved |
| use AgeTransGAN instead of SAM (ablation) | age transformation · FFHQ-Aging 20-29 to 0-2 | age mean error (years): 3.9 -> 0.4 | Improved |
| use AgeTransGAN instead of SAM | age transformation · CAF 20-29 to 7-9 | verification rate: 85.8 -> 99.7 | Improved |
| use rectified thresholds instead of common threshold 76.5 (ablation) | Face++ verification · FFHQ-Aging 20-29 to 0-2 | verification rate: 80.3(10.3) | Mixed |
Closest papers by meaning
- Towards Age-Invariant Face Recognition (TPAMI 2020 · similarity 0.90)
- Look across Elapse: Disentangled Representation Learning and Photorealistic Cross-Age Face Synthesis for Age-Invariant Face Recognition (AAAI 2019 · similarity 0.90)
- When Age-Invariant Face Recognition Meets Face Age Synthesis: A Multi-Task Learning Framework (CVPR 2021 · similarity 0.90)
- From Cradle to Cane: A Two-Pass Framework for High-Fidelity Lifespan Face Aging (NeurIPS 2025 · similarity 0.90)
- Face Aging With Identity-Preserved Conditional Generative Adversarial Networks (CVPR 2018 · similarity 0.89)
- A Call to Reflect on Evaluation Practices for Age Estimation: Comparative Analysis of the State-of-the-Art and a Unified Benchmark (CVPR 2024 · similarity 0.89)
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.
