MotioNet: 3D Human Motion Reconstruction from Monocular Video with Skeleton Consistency.
Mingyi ShiKfir AbermanAndreas AristidouTaku KomuraDani LischinskiDaniel Cohen-OrBaoquan ChenPublished in: CoRR (2020)
Keyphrases
- temporal consistency
- video data
- video sequences
- image sequences
- video streams
- real time
- image reconstruction
- three dimensional
- shape analysis
- multimedia
- video surveillance
- video content
- video frames
- pose estimation
- articulated human motion
- human motion
- space time
- video clips
- video analysis
- human motion tracking
- reconstruction method
- visual data
- monocular video sequences
- model based tracking
- ego motion
- medial axis
- key frames
- endpoints
- human activities