论文信息 - ARF @ MediaEval 2012: An Uninformed Approach to Violence Detection in Hollywood Movies

ARF @ MediaEval 2012: An Uninformed Approach to Violence Detection in Hollywood Movies

The MediaEval 2012 Aect Task challenged participants to automatically nd violent scenes in a set of Hollywood movies. We propose to rst predict a set of mid-level concept annotations from low-level visual and auditory features, then fuse the concept predictions and features to detect violent content. Instead of engineering features suitable for the task, we deliberately restrict ourselves to simple generalpurpose features with limited temporal context and a generic neural network classier,

[1] Cordelia Schmid,et al. Learning Color Names for Real-World Applications , 2009, IEEE Transactions on Image Processing.

[2] Chuan Liu,et al. Classification of Music and Speech in Mandarin News Broadcasts , 2007 .

[3] Nitish Srivastava,et al. Improving neural networks by preventing co-adaptation of feature detectors , 2012, ArXiv.

[4] Mohammad Soleymani,et al. The MediaEval 2011 Affect Task: Violent Scenes Detection in Hollywood movies , 2011, MediaEval.

[5] Patrick Lambert,et al. Improved Cut Detection for the Segmentation of Animation Movies , 2006, 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings.