Research map / Label-efficient and robust learning / Recognition and long-tail
Synthetic Data in Human Analysis: A Survey
Indu Joshi, Marcel Grimmer, Christian Rathgeb, Christoph Busch, François Brémond, Antitza Dantcheva
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 |
|---|---|---|---|
| replace real data with synthetic data | Human body segmentation model (Varol (body seg)) · Human body segmentation | Accuracy: 58.54 -> 56.51 | Hurt |
| add synthetic data to real training data | Action recognition model (Souza) · Action recognition | Accuracy: 93.3 -> 92.7 | Hurt |
| add synthetic data to real training data | People detection model (Aranjuelo) · People detection | Average Precision: 70 -> 82 | Improved |
| replace real data with synthetic data | Crowd counting model (Wang (crowd)) · Crowd counting | MAE: 275.5 -> 225.9 | Improved |
| add synthetic data to real training data | Unseen iris PAD model (Yadav) · Unseen iris PAD | EER: 25.18 -> 18.52 | Improved |
| add synthetic data to real training data | Speaker recognition model (Bird (avg)) · Speaker recognition | Average Accuracy: 95.48 -> 99.35 | Improved |
| replace real data with synthetic data | Gender classification from periocular images model (Tapia) · Gender classification from periocular images | Accuracy: 82.76 -> 91.9 | Improved |
| add synthetic data to real training data | Voice recognition model (Anton) · Voice recognition | Accuracy: 51.55 -> 57.76 | Improved |
| add synthetic data to real training data | Bot detection model (Acien) · Bot detection | EER: 23.2 -> 0.8 | Improved |
| add synthetic data to real training data | Face detection model (Han) · Face detection | Average Precision: 64 -> 74.5 | Improved |
| replace real data with synthetic data | Head pose estimation model (Basak) · Head pose estimation | MAE: 6.34 -> 5.13 | Improved |
| add synthetic data to real training data | Gait recognition model (Dou) · Gait recognition | Rank-1 Accuracy: 95.0 -> 96.4 | Improved |
Closest papers by meaning
- AugGen: Synthetic Augmentation using Diffusion Models Can Improve Recognition (NeurIPS 2025 · similarity 0.86)
- Real-Fake: Effective Training Data Synthesis Through Distribution Matching (ICLR 2024 · similarity 0.86)
- Alice Benchmarks: Connecting Real World Re-Identification with the Synthetic (ICLR 2024 · similarity 0.86)
- Improving the Scaling Laws of Synthetic Data with Deliberate Practice (ICML 2025 · similarity 0.86)
- Debiasing Synthetic Data Generated by Deep Generative Models (NeurIPS 2024 · similarity 0.85)
- Maximizing the Potential of Synthetic Data: Insights from Random Matrix Theory (ICLR 2025 · similarity 0.85)
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.
