As public video surveillance systems are everyday more common all over the world, the intereset towards solution to automatically detect violence and crimes in videos increases. In fact, taking decisions in a short time is essential in crime prevention, even when in need of watching hours of video footages.
In this regard, the candidate will work on the development and testing of deep learning algorithms to perform automatic violence detection in recorded videos as well as in live streamings. A specific focus will be given to end-to-end solutions, looking for interpretable results.
The thesis will be developed at the Information Engineering Department of the “Università Politecnica delle Marche”.
Involed tools and technologies: Keras, TensorFlow, Google Colab, Python, and similar.
- Accattoli, S., Sernani, P., Falcionelli, N., Mekuria, D. N., & Dragoni, A. F. (2020). Violence Detection in Videos by Combining 3D Convolutional Neural Networks and Support Vector Machines. Applied Artificial Intelligence, 34(4), 329-344.