Research map / Video / Action recognition
PrivHAR: Recognizing Human Actions from Privacy-Preserving Lens
Carlos Hinojosa, Miguel Marquez, Henry Arguello, Ehsan Adeli, Li Fei-Fei, Juan Carlos Niebles
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 adversarial training of the lens (ablation) | PrivHAR C3D · PA-HMDB51 | harmonic mean P: 38.6 -> 41.5 | Improved |
| add adversarial training of the lens (ablation) | PrivHAR RubiksNet · PA-HMDB51 | harmonic mean P: 40.9 -> 46.1 | Improved |
| add adversarial training of the lens (ablation) | PrivHAR C3D · PA-HMDB51 | action accuracy A_C: 51.1 -> 63.3 | Improved |
| add adversarial training of the lens (ablation) | PrivHAR RubiksNet · PA-HMDB51 | action accuracy A_C: 57.6 -> 73.8 | Improved |
| add temporal similarity matrix loss (ablation) | PrivHAR C3D · PA-HMDB51 | action accuracy A_C: 59.9 -> 63.3 | Improved |
| add temporal similarity matrix loss (ablation) | PrivHAR RubiksNet · PA-HMDB51 | action accuracy A_C: 72.3 -> 73.8 | Improved |
| increase Zernike coefficients from 50 to 200 (ablation) | PrivHAR C3D · PA-HMDB51 | action accuracy A_C: 58.3 -> 63.3 | Improved |
| increase Zernike coefficients from 50 to 200 (ablation) | PrivHAR RubiksNet · PA-HMDB51 | action accuracy A_C: 70.2 -> 73.8 | Improved |
| increase Zernike coefficients from 50 to 200 (ablation) | PrivHAR RubiksNet · PA-HMDB51 | SSIM (lower is more distortion): 0.618 -> 0.588 | Improved |
| use RubiksNet instead of C3D as HAR backbone (ablation) | PrivHAR · PA-HMDB51 | harmonic mean P: 41.7 -> 46.1 | Improved |
| PrivHAR-RubiksNet over lens of prior work with RubiksNet (ablation) | PrivHAR RubiksNet · PA-HMDB51 | action accuracy A_C: 52.4 -> 73.8 | Improved |
Closest papers by meaning
- Privacy-Preserving Deep Action Recognition: An Adversarial Learning Framework and A New Dataset (TPAMI 2020 · similarity 0.90)
- Towards Privacy-Preserving Visual Recognition via Adversarial Training: A Pilot Study (ECCV 2018 · similarity 0.90)
- STPrivacy: Spatio-Temporal Privacy-Preserving Action Recognition (ICCV 2023 · similarity 0.89)
- Selective Interpretable and Motion Consistent Privacy Attribute Obfuscation for Action Recognition (CVPR 2024 · similarity 0.86)
- EgoPrivacy: What Your First-Person Camera Says About You? (ICML 2025 · similarity 0.83)
- Validating Privacy-Preserving Face Recognition under a Minimum Assumption (CVPR 2024 · similarity 0.82)
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.
