SCTF: an efficient neural network based on local spatial compression and full temporal fusion for video violence detection - Zhenhua T, Zhenche X, Pengfei W, Danke W, Li L.

Spatiotemporal modeling is key for action recognition in videos. In this paper, we propose a Spatial features Compression and Temporal features Fusion (SCTF) block, including a Local Spatial features Compression (LSC) module and a Full Temporal features Fu...
Source: SafetyLit - Category: International Medicine & Public Health Tags: Media, Marketing, and Internet Issues Source Type: news