Agu, Emmanuel O.
Gait recognition using smartphone motion sensors such as accelerometers and gyroscopes is relatively underdeveloped compared to those using machine vision. This project explored the various state of the art neural networks-based approaches for accelerometer and gyroscope-based gait analysis and evaluated them. CNN and LSTM neural networks architectures proposed in prior work are replicated to achieve similar results on a gait dataset gathered in the wild. Prior work focused deep learning models for gait recognition on data gathered in controlled user studies.
Worcester Polytechnic Institute
Major Qualifying Project
All authors have granted to WPI a nonexclusive royalty-free license to distribute copies of the work, subject to other agreements. Copyright is held by the author or authors, with all rights reserved, unless otherwise noted.
Restricted-WPI community only
Available for download on Monday, May 16, 2022
Your accessibility may vary due to other restrictions.