PROJECT TITLE :
Gait Recognition Under Various Viewing Angles Based on Correlated Motion Regression
It is well recognized that gait is a crucial biometric feature to spot a person at a distance, e.g., in video surveillance application. But, in point of fact, amendment of viewing angle causes vital challenge for gait recognition. A unique approach using regression-primarily based view transformation model (VTM) is proposed to address this challenge. Gait options from across views can be normalized into a typical view using learned VTM(s). In principle, a VTM is used to remodel gait feature from one viewing angle (source) into another viewing angle (target). It consists of multiple regression processes to explore correlated walking motions, that are encoded in gait features, between supply and target views. In the learning processes, sparse regression based on the elastic net is adopted because the regression perform, that is free from the problem of overfitting and ends up in a lot of stable regression models for VTM construction. Based on widely adopted gait database, experimental results show that the proposed methodology significantly improves upon existing VTM-primarily based methods and outperforms most different baseline ways reported in the literature. Many sensible scenarios of applying the proposed methodology for gait recognition below various views also are mentioned during this paper.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here