• Shrayan Banerjee
• Vamshi Krishna Reddy

Neural Networks have been the at the core of solving complex problems. The goal of the project is to obtain a system that detects unwarranted and abusive behaviour from a video stream such as the one provided by the security camera. The neural network model utilizes the pose-net architecture to derive the results from the fed in images and then classifies them as violent or non violent according to the training we have provided and the system also have a method to take in new images to add to its database.

Deep Learning , Pose Estimation , Open pose , neural networks , partial affinity vectors , violence detection , surveillance

"A deep learning approach to predict the nature of security camera footage with the help of posenet architecture", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 4, page no.351-355, April-2019, Available :http://www.jetir.org/papers/JETIR1904554.pdf

"A deep learning approach to predict the nature of security camera footage with the help of posenet architecture", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 4, page no. pp351-355, April-2019, Available at : http://www.jetir.org/papers/JETIR1904554.pdf