GaitForeMer: Self-Supervised Pre-Training of Transformers via Human Motion Forecasting for Few-Shot Gait Impairment Severity Estimation.
Mark EndoKathleen L. PostonEdith V. SullivanLi Fei-FeiKilian M. PohlEhsan AdeliPublished in: CoRR (2022)
Keyphrases
- human motion
- gait recognition
- motion capture
- video sequences
- gait analysis
- human movement
- human body
- temporal segmentation
- optical motion capture
- body parts
- image sequences
- human motion tracking
- dynamical model
- spatio temporal
- human actions
- motion recognition
- motion sequences
- human motion analysis
- markerless
- human movements
- motion estimation
- articulated motion
- human identification
- motion capture data
- object recognition
- visual data
- human pose
- training set
- machine learning
- motion synthesis
- motion analysis
- video segmentation
- person identification
- articulated objects
- video shots
- humanoid robot
- video frames
- camera motion
- action recognition
- pose estimation
- video data
- viewpoint
- moving objects
- three dimensional