Abstract: Skeletal graphs can represent concise and reliable features for human-activity recognition in recent years. However, they have to be acquired by Kinect sensors or regular cameras, which rely ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results