You2Me: Inferring Body Pose in Egocentric Video via First and Second Person Interactions.
Evonne NgDonglai XiangHanbyul JooKristen GraumanPublished in: CVPR (2020)
Keyphrases
- body pose
- human body
- human actions
- body parts
- video sequences
- video frames
- video data
- activity recognition
- pose estimation
- appearance model
- markerless
- human pose estimation
- motion capture
- multimedia
- globally optimal
- action recognition
- prior model
- human pose
- human motion
- generative model
- spatio temporal
- single camera
- spatial and temporal
- motion tracking
- multibody
- space time
- background subtraction
- human activities
- low resolution