details
Human activity recognition (HAR) is an important and challenging task in the area of computer vision research. The goal of HAR is to automatically analyze and detect ongoing activities from video. Detecting specific activities in a live feed or searching in video archives still relies almost completely on human resources. Detecting multiple activities in real-time video feeds is currently performed by assigning multiple analysts to simultaneously watch the same video stream. Manual analysis of video is labour intensive, fatiguing, and error prone. This area has grown dramatically in the past 10 years, and has several applications like surveillance systems, human computer interfaces, sports video analysis, digital shopping assistants, video retrieval, gaming and health-care. The next stage of research would be to infer the future actions of people from visual input. We propose to expand the current vision-based activity analysis to a level where it is possible to predict the future actions executed by a subject.
The summarized objectives are:
Detecting relevant human behaviour in midst of irrelevant additional motion;
Recognizing the detected actions among several pre-learned actions;
Given the current recognized action, predicting the next most likely action or behaviour that will occur in a near future.
