You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Violence recognition in streaming video using Transfer Learning and MoViNets. The project leverages state-of-the-art deep learning techniques to create an efficient and accurate violence detection system.
My main goal is to build an innovative technological solution that could detect sexual harassment in real-time, which is much needed to fill in the gap left void by traditional methods, by which I aim to create a safer work environment, overcome reporting hesitancy, support HR and legal management, reduce psychological and physical stress.
Child labour is a significant problem in most developing countries, damaging, spoiling and destroying children's futures. Child labour can result in extreme bodily and mental harm and even death. To overcome the problem, we have approached a technical solution by creating a model using video classification techniques to detect child labour actvts.
Python application for real-time multi-camera CCTV video ingestion and person detection using TensorFlow's SSD MobileNet V2. Features parallel processing, robust error handling, and saves processed videos in MP4 format.