Christine Perez edited Human_Activity_recognition_has_been__.tex  about 8 years ago

Commit id: 3083f388a326197745252a47e9de0904dd1cf5a0

deletions | additions      

       

Human Activity recognition has been a critical range of Computer vision research subsequent to the 1980s. The previous decade has seen a fast advancement of 3D data acquisition techniques \cite{aggarwal2014human} \cite{brandvik2012investigating}. Such techniques involve recent advances in 3d depth cameras and motion capture technologies mostly for video games control interfaces that have brought motion capture into high extent \cite{regazzoni2014rgb}. According to \citet{brandvik2012investigating}, in order to increase the reasoning capacity of each computer the use of machine learning techniques which are a large part of Artificial intelligence research should be applied. It is prominent in the Kinect motion capture technology that the ability to predict logical outcomes by information acquired from data sets that can contains patterns and prediction of correlated patterns that ca recognize objects in images are solely an application of machine learning.  In the work of \citet{zhang2012microsoft}, he stated that the Kinect sensor incorporates several advanced sensing hardware suitable for a wide range of applications that may vary from entertainment domain (e.g., video games, virtual characters in movies) to the Bio-mechanical and biomedical domain (e.g., gait analysis or orthopedic rehabilitation) and to a huge number of industrial sectors \cite{regazzoni2014rgb}. Most notably, it contains a depth sensor, a color camera, and a four-microphone array that provide full-body 3D motion capture, facial recognition, and voice recognition capabilities \cite{smisek20133d}. The depth sensor portion of the Kinect consists of the IR projector combined with the IR camera, which is a monochrome complementary metal oxide semiconductor (CMOS) sensor an a depth sensing technology based on Structured light principle.