Vid2Avatar: 3D Avatar Reconstruction from Videos in the Wild via Self-supervised Scene Decomposition.
Chen GuoTianjian JiangXu ChenJie SongOtmar HilligesPublished in: CVPR (2023)
Keyphrases
- video sequences
- virtual world
- three dimensional
- human computer interaction
- dynamic scenes
- sign language
- virtual reality
- image reconstruction
- moving camera
- image sequences
- traffic scenes
- visual data
- piecewise planar
- video content
- human motion
- d scene
- single image
- video sequences captured
- neighboring frames
- high resolution
- stereo reconstruction
- uncalibrated cameras
- motion features
- video images
- reconstruction method
- real scenes
- facial animation
- input video
- moving objects
- video frames
- architectural scenes
- multiple objects