Research map / Label-efficient and robust learning / Recognition and long-tail
TimeBooth: Disentangled Facial Invariant Representation for Diverse and Personalized Face Aging
Zepeng Su, Zhulin Liu, Zongyan Zhang, Tong Zhang, C.L.Philip Chen
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 Shuffle Strategy (swap ID feature with same identity at another age) (ablation) | TimeBooth · Ablation | R-AgeMAE: 8.62 -> 4.58 | Improved |
| add Dynamic Training-set Sampling (DTS) (ablation) | TimeBooth · Ablation | R-AgeMAE: 8.37 -> 4.58 | Improved |
| add Dynamic Training-set Sampling (DTS) (ablation) | TimeBooth · Ablation | KID (x100): 7.40 -> 4.25 | Improved |
| add Age Prototype Replacement (PAR/APR) (ablation) | TimeBooth · Ablation | KID (x100): 6.80 -> 4.25 | Improved |
| add Age Prototype Replacement (PAR/APR) (ablation) | TimeBooth · Ablation | R-AgeMAE: 4.11 -> 4.58 | Hurt |
| add Age Prototype Replacement (PAR/APR) (ablation) | TimeBooth · Ablation | Face Sim: 0.719 -> 0.733 | Improved |
| add Shuffle Strategy (ablation) | TimeBooth · Ablation | KID (x100): 4.85 -> 4.25 | Improved |
| TimeBooth vs DreamBooth (tuning-free image-prompt age vs per-subject fine-tuning) | TimeBooth · Young-to-Old | R-AgeMAE: 5.76 -> 3.53 | Improved |
| TimeBooth vs DreamBooth on young-to-child | TimeBooth · Young-to-Child | R-AgeMAE: 5.65 -> 7.97 | Hurt |
| increase aging guidance alpha toward Age Prompt (ablation) | TimeBooth · Qualitative (Fig. 11) | age transfer vs fidelity: best balance at alpha 0.7 to 0.9 | Mixed |
Closest papers by meaning
- Hierarchical Face Aging through Disentangled Latent Characteristics (ECCV 2020 · similarity 0.93)
- Disentangled Representation for Age-Invariant Face Recognition: A Mutual Information Minimization Perspective (ICCV 2021 · similarity 0.93)
- Look across Elapse: Disentangled Representation Learning and Photorealistic Cross-Age Face Synthesis for Age-Invariant Face Recognition (AAAI 2019 · similarity 0.91)
- Towards Age-Invariant Face Recognition (TPAMI 2020 · similarity 0.91)
- Temporal Non-Volume Preserving Approach to Facial Age-Progression and Age-Invariant Face Recognition (ICCV 2017 · similarity 0.90)
- From Cradle to Cane: A Two-Pass Framework for High-Fidelity Lifespan Face Aging (NeurIPS 2025 · similarity 0.90)
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.
