Video Violence Detection on DVD Dataset
74.13Top-1 AccuracyFuseMamba-VD
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| FuseMamba-VDArchitecture=SSM, Pretraining=K-400, Input Size=64×224², Params. (M)=154.3, FLOPs (G)=19122025.05 | 74.13 | 68.97 | 77.82 | |
| CUE-NetArchitecture=CNN+Trans., Pretraining=K-400, Input Size=64×224², Params. (M)=354, FLOPs (G)=59062025.05 | 73.68 | 68.71 | 77.28 | |
| VideoMamba-MArchitecture=SSM, Pretraining=K-400, Input Size=64×224², Params. (M)=74, FLOPs (G)=8062025.05 | 72.47 | 66.42 | 76.67 | |
| VideoMambaPro-MArchitecture=SSM, Pretraining=K-400, Input Size=64×224², Params. (M)=76, FLOPs (G)=10282025.05 | 72.01 | 68.05 | 75.1 | |
| VideoMamba-LArchitecture=SSM, Pretraining=K-400, Input Size=64×224², Params. (M)=148, FLOPs (G)=16442025.05 | 71.1 | 63.76 | 75.97 | |
| Uniformer-V2Architecture=CNN+Trans., Pretraining=K-400, Input Size=64×224², Params. (M)=354, FLOPs (G)=61082025.05 | 70.95 | 63.36 | 75.94 | |
| VideoSwin-BArchitecture=Trans., Pretraining=K-400, Input Size=64×224², Params. (M)=88, FLOPs (G)=281.62025.05 | 64.3 | 60.4 | 67.49 | |
| SlowFastArchitecture=CNN, Pretraining=-, Input Size=64×224², Params. (M)=60, FLOPs (G)=2342025.05 | 61.72 | 55.85 | 66.22 |