Application of video frame interpolation to markerless, single-camera gait analysis

Marcus Dunn*, Adam Kennerley, Zhane Murrell-Smith, Kate Webster, Kane Middleton, Jon Wheat

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)
31 Downloads (Pure)


In clinic settings, factors such as time, cost, expertise, and technology feasibility limit the use of instrumented biomechanical analysis. Recent advances in commercial markerless motion capture systems can address patient ease-of-use factors, but are high cost and require specialised equipment, dedicated spaces, and technical expertise. As such, they present similar limitations to biomechanical analyses in clinic settings. Single-camera pose estimation techniques have generated cautious optimism for markerless gait analysis. However, parameters derived using low-cost and low-sample rate cameras commonly used in clinic settings are not yet accurate enough to detect change in complex movement systems. Video frame interpolation is a single-step process that artificially increases the sample rate of videos. This study applied video frame interpolation to videos of walking and demonstrates improved precision for step, stance, swing and double support times, as well as marginal improvements to the precision of ankle and knee joint angles, derived by single-camera pose estimation. Video frame interpolation potentially represents a delimiting factor for gait analysis in clinic settings, as limiting factors such as time, cost, technology feasibility and patient ease-of-use can be minimised.

Original languageEnglish
Article number22
JournalSports Engineering
Publication statusPublished - 28 Apr 2023

ASJC Scopus subject areas

  • Biomedical Engineering
  • Modelling and Simulation
  • Orthopedics and Sports Medicine
  • Physical Therapy, Sports Therapy and Rehabilitation
  • Mechanics of Materials
  • Mechanical Engineering


Dive into the research topics of 'Application of video frame interpolation to markerless, single-camera gait analysis'. Together they form a unique fingerprint.

Cite this