Research map / Video / Action recognition
Bullying10K: A Large-Scale Neuromorphic Dataset towards Privacy-Preserving Bullying Recognition
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 |
|---|---|---|---|
| Increase frame interval and temporal step length | Action recognition models on Bullying10K · Bullying10K action recognition | precision | Helped |
| Use DVS input instead of RGB input | Action recognition models · Bullying10K action recognition | accuracy: 64.00 -> 66.80; 59.25 -> 69.00; 63.20 -> 70.80 | Helped |
| Apply RGB privacy-protection methods rather than using unmodified RGB | R3D, SlowFast, and X3D · Bullying10K action recognition | accuracy: 64.00 -> 63.30; 59.25 -> 57.80; 63.20 -> 60.75 | Hurt |
| Use DVS rather than the listed RGB privacy filters | R3D, SlowFast, and X3D · Bullying10K action recognition | accuracy: 66.80 vs 63.30; 69.00 vs 57.80; 70.80 vs 60.75 | Helped |
| Use TSM rather than TSN features | BMN · Bullying10K temporal action localization | AUC: 80.3 -> 81.0 | Helped |
| Use TSM rather than TSN features | BSN · Bullying10K temporal action localization | AUC: 75.9 -> 75.5 | Hurt |
| Replace the ResNet-101 backbone with a spiking ResNet-50 backbone | SimpleBaseline · Bullying10K pose estimation | AP: 63.6 -> 54.1 | Hurt |
| Use HRNet-ws32 rather than SimpleBaseline with ResNet-50 | Pose estimation · Bullying10K pose estimation | AP: 62.6 -> 62.7 | No clear effect |
Closest papers by meaning
- Beyond Euclidean: Dual-Space Representation Learning for Weakly Supervised Video Violence Detection (NeurIPS 2024 · similarity 0.84)
- VPD-100K: Towards Generalizable and Fine-grained Visual Privacy Protection (ICML 2026 · similarity 0.83)
- HARDVS: Revisiting Human Activity Recognition with Dynamic Vision Sensors (AAAI 2024 · similarity 0.82)
- UCF-Crime-DVS: A Novel Event-Based Dataset for Video Anomaly Detection with Spiking Neural Networks (AAAI 2025 · similarity 0.81)
- Selective Interpretable and Motion Consistent Privacy Attribute Obfuscation for Action Recognition (CVPR 2024 · similarity 0.80)
- Image Safeguarding: Reasoning with Conditional Vision Language Model and Obfuscating Unsafe Content Counterfactually (AAAI 2024 · similarity 0.80)
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.