Research map / Federated learning and privacy / Privacy attacks and secure learning
Privacy-centric Deep Motion Retargeting for Anonymization of Skeleton-Based Motion Visualization
Thomas Carr, Depeng Xu, Shuhan Yuan, Aidong Lu
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 embedding classifiers M and P (ablation) | PMR · NTU-60 (CV, constant dummy) | Re-ID: 0.1575 -> 0.0778 | Improved |
| add embedding classifiers M and P (ablation) | PMR · NTU-60 (CV, constant dummy) | Action recognition: 0.3753 -> 0.3560 | Improved |
| add smoothness loss (ablation) | PMR · NTU-60 (CV, constant dummy) | Re-ID: 0.1361 -> 0.0778 | Improved |
| add smoothness loss (ablation) | PMR · NTU-60 (CV, constant dummy) | MSE: 0.0117 -> 0.0138 | Improved |
| add latent consistency loss (ablation) | PMR · NTU-60 (CV, constant dummy) | Re-ID: 0.1260 -> 0.0778 | Improved |
| add quality controller Q (ablation) | PMR · NTU-60 (CV, constant dummy) | Re-ID: 0.0930 -> 0.0778 | Improved |
| add quality controller Q (ablation) | PMR · NTU-60 (CV, constant dummy) | Action recognition: 0.2961 -> 0.3560 | Improved |
| increase alpha_emb from 0 to 0.5 (ablation) | PMR · NTU-60 (CV, constant dummy) | Re-ID Top-1: 0.1926 -> 0.0684 | Improved |
| increase alpha_emb from 0.5 to 40 (ablation) | PMR · NTU-60 (CV, constant dummy) | MSE: 0.0110 -> 0.0261 | Hurt |
| increase alpha_emb from 0.5 to 40 (ablation) | PMR · NTU-60 (CV, constant dummy) | Re-ID Top-1: 0.0684 -> 0.0559 | Improved |
| use random instead of constant dummy | PMR · NTU-60 (CV) | Action recognition Top-1: 0.3560 -> 0.3049 | Hurt |
| use random instead of constant dummy | PMR · NTU-60 (CV) | Re-ID Top-1: 0.0778 -> 0.0647 | Improved |
Closest papers by meaning
- DisentangledTMR: Privacy-Preserving Skeleton Motion Retargeting via Factorized Transformers (ECCV 2026 · similarity 0.88)
- Anonymization for Skeleton Action Recognition (AAAI 2023 · similarity 0.86)
- Vulnerability of Privacy-Preserving Visual Localization against Diffusion-based Attacks (ECCV 2026 · similarity 0.84)
- ReGenHuman: Re-Generating Human Appearances for Realistic Full-Body Video Anonymization (NeurIPS 2026 · similarity 0.83)
- Privacy-Preserving Deep Action Recognition: An Adversarial Learning Framework and A New Dataset (TPAMI 2020 · similarity 0.83)
- EgoPrivacy: What Your First-Person Camera Says About You? (ICML 2025 · similarity 0.81)
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.
