Filling the Gaps: Predicting Missing Joints of Human Poses Using Denoising Autoencoders.
Nicolò CarissimiPaolo RotaCigdem BeyanVittorio MurinoPublished in: ECCV Workshops (2) (2018)
Keyphrases
- denoising
- human pose
- human body
- body parts
- human motion
- pose estimation
- monocular images
- image denoising
- action recognition
- natural images
- human detection
- image processing
- pedestrian detection
- degrees of freedom
- video sequences
- human actions
- spatial and temporal
- computer vision
- machine learning
- depth images
- articulated objects
- body pose
- human object interactions
- latent variables
- video frames
- topic models
- maximum likelihood
- higher order
- probabilistic model