Video summarization (action recognition)

Abstract

This is a report for the Final Year Project held in the Academic Year 2010/2011. The Final Year Project is being held over the last 2 semester of Electrical and Electronics Engineering course (EEE) at the Final Year of Nanyang Technological University (NTU). Researching into the Final Year Project, in the past decade, the industry has witnessed a rapid proliferation of video cameras in all walks of life and had resulted in a tremendous explosion of video content. Several applications such as content-based video annotation and retrieval, highlight extraction and video summarization requires recognition of the human activities occurring in the videos. The analysis of human activities in a video is an area with increasingly important consequences from security and surveillance to entertainment and personal archiving. Therefore the potential for researching into Action Recognition was realized.Corner detection is used as the first step of many vision tasks such as tracking, SLAM (simultaneous localisation and mapping), localisation, image matching and recognition. Hence, a large number of corner detectors exist in the literature. With so many already available detectors in the market, it may appear unnecessary to present yet another detector to the community; however, we have a strong interest in real-time frame rate applications such as SLAM in which computational resources are at a premium. In this report, there will be several comparisons of the feature detectors, the aim being targeted at finding the best feature detector for this Project and incorporating the corner detector into the Graphical User Interface System (GUI).