Research map / Label-efficient and robust learning / Recognition and long-tail
Personalized Age Progression with Bi-Level Aging Dictionary Learning
Xiangbo Shu, Jinhui Tang, Zechao Li, Hanjiang Lai, Liyan Zhang, Shuicheng Yan
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 |
|---|---|---|---|
| synthesize aging faces with BDL-PAP before verification | face verification · FG-NET cross-age verification | EER (%): 14.89 -> 8.06 | Improved |
| use BDL-PAP instead of IAAP to age younger faces | face verification · FG-NET cross-age verification | EER (%): 10.36 -> 8.06 | Improved |
| use BDL-PAP instead of RFA | face verification · FG-NET cross-age verification | EER (%): 8.69 -> 8.06 | Improved |
| use BDL-PAP instead of CDL-PAP | face verification · FG-NET cross-age verification | EER (%): 8.53 -> 8.06 | Improved |
| add personalized layer to aging dictionary model (ablation) | BDL-AP · FG-NET user study | average rating score: 0.4221 -> 1.5779 | Improved |
| compare BDL-PAP with CDL-PAP in user votes | CDL-PAP · 261 comparison groups | % of votes as best: 34.8 -> 36.5 | No clear effect |
| use BDL-PAP rather than prior works in user votes | prior works · 261 comparison groups | % of votes as best: 26.7 -> 36.5 | Improved |
| use bi-level dictionary (BDL-PAP) rather than coupled dictionary (CDL-PAP) for speed, 0-5 to 6-10 | CDL-PAP · age progression synthesis | running time (s): 13.5 -> 2.2 | Improved |
| use BDL-PAP rather than CDL-PAP for speed, 0-5 to 61-80 | CDL-PAP · age progression synthesis | running time (s): 108.0 -> 17.6 | Improved |
| use BDL-PAP rather than IAAP for speed | IAAP · age progression sequence | running time (s): 72.0 -> 17.6 | Improved |
Closest papers by meaning
- Longitudinal Face Modeling via Temporal Deep Restricted Boltzmann Machines (CVPR 2016 · similarity 0.89)
- Recurrent Face Aging (CVPR 2016 · similarity 0.88)
- TimeBooth: Disentangled Facial Invariant Representation for Diverse and Personalized Face Aging (ICCV 2025 · similarity 0.88)
- Temporal Non-Volume Preserving Approach to Facial Age-Progression and Age-Invariant Face Recognition (ICCV 2017 · similarity 0.88)
- Learning Continuous Face Age Progression: A Pyramid of GANs (TPAMI 2019 · similarity 0.87)
- From Cradle to Cane: A Two-Pass Framework for High-Fidelity Lifespan Face Aging (NeurIPS 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.
