Vid2Avatar: 3D Avatar Reconstruction from Videos in the Wild via Self-supervised Scene Decomposition.
Chen GuoTianjian JiangXu ChenJie SongOtmar HilligesPublished in: CoRR (2023)
Keyphrases
- virtual world
- video sequences
- three dimensional
- human computer interaction
- piecewise planar
- dynamic scenes
- virtual reality
- video images
- sign language
- d scene
- single image
- stereo reconstruction
- input video
- traffic scenes
- scene reconstruction
- urban scenes
- motion features
- moving camera
- neighboring frames
- high resolution
- video content
- image reconstruction
- video frames
- computer vision
- object recognition
- uncalibrated cameras
- dynamic textures
- scene analysis
- surveillance videos
- human activities
- multiple images
- video scene
- scene understanding
- space time
- visual features
- static camera
- ground plane
- moving objects
- complex scenes
- video clips