MeshAvatar: Learning High-quality Triangular Human Avatars from Multi-view Videos.
Yushuo ChenZerong ZhengZhe LiChao XuYebin LiuPublished in: CoRR (2024)
Keyphrases
- multi view
- depth map
- high quality
- multi view learning
- multiple views
- single view
- three dimensional
- learning algorithm
- camera calibration
- dynamic scenes
- co training
- multiple cameras
- learning process
- d objects
- surface reconstruction
- supervised learning
- semi supervised
- active learning
- multi view stereo
- view invariant
- learning problems
- video frames
- range images
- virtual world
- semi supervised learning
- unsupervised learning
- image quality
- high resolution
- video sequences