Gait Estimation and Analysis from Noisy Observations

EasyChair Preprint 295

Abstract

People's walking style - gait - can be an indicator of their health as it is affected by pain, illness, weakness, and aging. Gait analysis aims to detect gait variations. It is usually performed by an experienced observer with the help of cameras, sensors, or other devices. Frequent gait analysis, to observe changes over time, is costly and impractical. Here, we first discuss estimating gait movements from predicted 2D joint locations that represent selected body parts from videos. Then, we use a long-short term memory (LSTM) regression model to predict 3D (Vicon) data, which was recorded simultaneously with the videos as ground truth. Feet movements estimated from video correlate highly with the Vicon data, enabling gait analysis by measuring selected spatial gait parameters (step and cadence length, and walk base) from estimated movements. Using inexpensive and reliable cameras to record, estimate and analyse a person's gait can be helpful; early detection of its changes facilitates early intervention.

Keyphrases: Regression, gait analysis, machine learning, vision data

@booklet{EasyChair:295,
  author    = {Hafsa Ismail and Ibrahim Radwan and Hanna Suominen and Roland Goecke},
  title     = {Gait Estimation and Analysis from Noisy Observations},
  doi       = {10.29007/57cc},
  howpublished = {EasyChair Preprint 295},
  year      = {EasyChair, 2018}}