Using CNN Algorithm Detection of Suspicious Activity from Video Surveillance

Abstract

Abstract: Detecting suspicious human activity involves predicting the positions of body parts or joints from an image or video. This problem has been extensively studied in computer vision for over 15 years. It is crucial because there are numerous applications that can benefit from activity detection. For instance, human pose estimation is used in various areas such as video surveillance, tracking and understanding animal behaviour, detecting sign language, improving human-computer interaction, and capturing motion without markers.