Research map / 3D vision / Human pose estimation
Jointly Optimize Data Augmentation and Network Training: Adversarial Data Augmentation in Human Pose Estimation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 adversarial scaling and rotating (ablation) | Residual hourglass · MPII validation | PCKh@0.5 mean: 87.0 -> 87.5 | Helped |
| Add adversarial hierarchical occluding (ablation) | Residual hourglass · MPII validation | PCKh@0.5 mean: 87.0 -> 87.4 | Helped |
| Add adversarial scaling and rotating plus hierarchical occluding (ablation) | Residual hourglass · MPII validation | PCKh@0.5 mean: 87.0 -> 87.8 | Helped |
| Add adversarial scaling and rotating (ablation) | Dense hourglass · MPII validation | PCKh@0.5 mean: 86.8 -> 87.3 | Helped |
| Add adversarial hierarchical occluding (ablation) | Dense hourglass · MPII validation | PCKh@0.5 mean: 86.8 -> 87.2 | Helped |
| Add adversarial scaling and rotating plus hierarchical occluding (ablation) | Dense hourglass · MPII validation | PCKh@0.5 mean: 86.8 -> 87.6 | Helped |
| Jointly train with adversarial scaling, rotating, and hierarchical occluding | Stacked HGs(8) · MPII test | PCKh@0.5 mean: 90.9 -> 91.5 | Helped |
| Jointly train with adversarial scaling, rotating, and hierarchical occluding | Stacked HGs(8) · LSP | PCK@0.2 mean: 93.0 -> 94.5 | Helped |
| Use adversarial scaling and rotating plus hierarchical occluding instead of either component alone (ablation) | Residual and dense hourglasses · MPII validation | PCKh@0.5 mean: 0.4% | Helped |
Closest papers by meaning
- A Dual-Augmentor Framework for Domain Generalization in 3D Human Pose Estimation (CVPR 2024 · similarity 0.90)
- Efficient and Effective Augmentation Strategy for Adversarial Training (NeurIPS 2022 · similarity 0.89)
- AugMax: Adversarial Composition of Random Augmentations for Robust Training (NeurIPS 2021 · similarity 0.87)
- A Bayesian Data Augmentation Approach for Learning Deep Models (NeurIPS 2017 · similarity 0.87)
- Adversarial AutoAugment (ICLR 2020 · similarity 0.87)
- When Human Pose Estimation Meets Robustness: Adversarial Algorithms and Benchmarks (CVPR 2021 · similarity 0.87)
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.