Research map / Vision-language and multimodal / Cross-modal learning
Not only Look, but also Listen: Learning Multimodal Violence Detection under Weak Supervision
Peng Wu, Jing Liu, Yujia Shi, Yujia Sun, Fangtao Shao, Zhaoyang Wu, Zhiwei Yang
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 audio (VGGish) to RGB input (ablation) | HL-Net · XD-Violence | AP (%): +3.2% | Improved |
| add audio (VGGish) to Flow input (ablation) | HL-Net · XD-Violence | AP (%): +3.5% | Improved |
| add localized and score branches to holistic branch (ablation) | HL-Net · XD-Violence | AP (%): 75.44 -> 78.64 | Improved |
| add holistic and score branches to localized branch (ablation) | HL-Net · XD-Violence | AP (%): 76.60 -> 78.64 | Improved |
| add score branch to holistic and localized branches (ablation) | HL-Net · XD-Violence | AP (%): 77.23 -> 78.64 | Improved |
| use offline detection instead of online (HLC approximator) | HL-Net · XD-Violence | AP (%): 73.67 -> 78.64 | Improved |
| use I3D RGB features instead of C3D | HL-Net · XD-Violence | AP (%): 67.19 -> 78.64 | Improved |
| use similarity Version 2 with learnable weights instead of Version 1 (ablation) | HL-Net · XD-Violence | AP (%): 78.64 -> 79.04 | No clear effect |
| increase sampling length threshold Gamma from 100 to 200 (ablation) | HL-Net · XD-Violence | AP (%): 78.30 -> 78.64 | No clear effect |
| add audio to RGB for Abuse class (ablation) | HL-Net · XD-Violence Abuse | AP (%): 90.10 -> 83.43 | Hurt |
| add audio to RGB for Shooting class (ablation) | HL-Net · XD-Violence Shooting | AP (%): 83.51 -> 90.30 | Improved |
Closest papers by meaning
- Graph Distillation for Action Detection with Privileged Modalities (ECCV 2018 · similarity 0.84)
- Multimodal Multipart Learning for Action Recognition in Depth Videos (TPAMI 2016 · similarity 0.83)
- PiercingEye: Dual-Space Video Violence Detection With Hyperbolic Vision-Language Guidance (TPAMI 2025 · similarity 0.83)
- There Is More Than Meets the Eye: Self-Supervised Multi-Object Detection and Tracking With Sound by Distilling Multimodal Knowledge (CVPR 2021 · similarity 0.83)
- Beyond Euclidean: Dual-Space Representation Learning for Weakly Supervised Video Violence Detection (NeurIPS 2024 · similarity 0.83)
- Vision + X: A Survey on Multimodal Learning in the Light of Data (TPAMI 2024 · similarity 0.83)
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.
