A dual-branch VideoMamba with gated class-token fusion achieves 95.85% accuracy on a newly combined violence-detection benchmark and 74.13% on DVD, with about half the parameters and FLOPs of the CUE-Net baseline.
Data efficient video transformer for violence detection
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
FuseMamba-VD: Dual Branch VideoMamba with Gated Class Token Fusion for Violence Detection
A dual-branch VideoMamba with gated class-token fusion achieves 95.85% accuracy on a newly combined violence-detection benchmark and 74.13% on DVD, with about half the parameters and FLOPs of the CUE-Net baseline.