Motion Projection Consistency-Based 3-D Human Pose Estimation With Virtual Bones From Monocular Videos.
Guangming WangHonghao ZengZiliang WangZhe LiuHesheng WangPublished in: IEEE Trans. Cogn. Dev. Syst. (2023)
Keyphrases
- image sequences
- independently moving objects
- view invariant
- moving camera
- optical flow
- motion features
- video sequences
- periodic motion
- camera motion
- ego motion
- image motion
- human motion
- body parts
- motion estimation
- input video
- action recognition
- motion trajectories
- recognition of human actions
- key frames
- motion analysis
- motion parameters
- space time
- temporal consistency
- successive frames
- video stabilization
- video data
- dynamic scenes
- motion capture data
- motion model
- static images
- pose estimation
- motion patterns
- perspective images
- projective geometry
- object motion
- multiple visual cues
- human actions
- video frames
- moving objects
- traffic scenes
- action detection
- spatial and temporal
- motion field
- dynamic textures
- static scene
- motion cues
- depth cues
- field of view
- motion segmentation
- human body tracking
- relative depth
- spatio temporal
- visual cues
- feature points
- computer vision
- visual features
- visual odometry
- x ray images
- visual data
- multibody
- global motion
- motion capture
- stereo camera