An Optimization Framework to Enforce Multi-View Consistency for Texturing 3D Meshes Using Pre-Trained Text-to-Image Models.
Zhengyi ZhaoChen SongXiaodong GuYuan DongQi ZuoWeihao YuanZilong DongLiefeng BoQixing HuangPublished in: CoRR (2024)
Keyphrases
- multi view
- pre trained
- d mesh
- d objects
- single view
- multiple views
- input image
- three dimensional
- training data
- shape descriptors
- depth map
- single image
- camera calibration
- semi supervised
- computer vision
- surface reconstruction
- high resolution
- similarity measure
- texture mapping
- surface model
- training examples
- feature points
- medical images
- image segmentation