Multimodal Violence Detection on NTU-CCTV (test)
88.63AccuracyCoLoRSMamba
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| CoLoRSMambaArchitecture=SSM, Modality=audio-video2026.04 | 88.63 | 86.24 | 90.31 | 88.28 | |
| LAVISHArchitecture=Transformer, Modality=audio-video2026.04 | 86.44 | 82.25 | 89.03 | 85.64 | |
| MBTArchitecture=Transformer, Modality=audio-video2026.04 | 85.26 | 82.83 | 87.08 | 84.96 | |
| CAV-MAE SyncArchitecture=Transformer, Modality=audio-video2026.04 | 84.92 | 82.12 | 86.96 | 84.54 | |
| DBVideoMambaArchitecture=SSM, Modality=video-only2026.04 | 83.99 | 81.45 | 85.93 | 83.69 | |
| TIMArchitecture=Transformer, Modality=audio-video2026.04 | 83.07 | 80.24 | 85.19 | 82.72 | |
| VideoMamba-MArchitecture=SSM, Modality=video-only2026.04 | 82.22 | 79 | 84.59 | 81.8 | |
| CUE-NetArchitecture=CNN+Transformer, Modality=video-only2026.04 | 82.06 | 79.18 | 84.23 | 81.71 | |
| VideoMambaPro-MArchitecture=SSM, Modality=video-only2026.04 | 80.96 | 77.45 | 83.53 | 80.49 | |
| UniFormer-V2Architecture=CNN+Transformer, Modality=video-only2026.04 | 80.37 | 77.18 | 82.78 | 79.98 | |
| SlowFast-AVArchitecture=CNN, Modality=audio-video2026.04 | 79.11 | 75.88 | 81.58 | 78.73 | |
| VideoSwin-BArchitecture=Transformer, Modality=video-only2026.04 | 78.77 | 75.49 | 81.28 | 78.39 | |
| DepMambaArchitecture=SSM, Modality=audio-video2026.04 | 78.18 | 74.68 | 80.83 | 77.76 | |
| SlowFastArchitecture=CNN, Modality=video-only2026.04 | 76.41 | 72.76 | 79.2 | 75.98 | |
| AudioMambaArchitecture=SSM, Modality=audio-only2026.04 | 72.79 | 68.55 | 76.02 | 72.29 | |
| ASTArchitecture=Transformer, Modality=audio-only2026.04 | 71.02 | 65.39 | 75.07 | 70.23 | |
| Resnet-50Architecture=CNN, Modality=audio-only2026.04 | 69.17 | 63.03 | 73.55 | 68.29 |