Human activity recognition

• Main Author: Ng, Kah Meng.
• Other Authors: Chua Chin Seng
• Format: Final Year Project
• Language: English
• Published: 2011
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Electronic systems::Signal processing
• Online Access: http://hdl.handle.net/10356/45589
• Institution: Nanyang Technological University

The human body is capable of a wide range of motion and with flexible limbs, a human cannot be considered as a rigid body. This makes for a challenging obstacle when one tries to track an individual for human activity recognition. For human activity recognition, the motion of individual limbs, torso or head of the person to be tracked are important and cannot be taken as a whole. As a result this brings about an issue of over-segmentation of the human body and therefore a greater need to track the individual blobs of representing the human body over the video sequence. Tracking of multiple human for activity recognition only brings about more problems such as occlusion. Other problems faced in the tracking of human in a video sequence involve lighting, recognizing 3D action in a 2D image. This paper aims to look at the tracking of the multiple individuals interacting in a video frame and tackle the issue of tracking for human activity recognition.