PIFu for the Real World: A Self-supervised Framework to Reconstruct Dressed Human from Single-view Images.
Zhangyang XiongDong DuYushuang WuJingqi DongDi KangLinchao BaoXiaoguang HanPublished in: CoRR (2022)
Keyphrases
- single view
- polyhedral objects
- three dimensional
- real world
- multi view
- input image
- man made objects
- vanishing points
- image data
- multiple views
- image retrieval
- object recognition
- perspective projection
- co training
- calibrated cameras
- surface reconstruction
- feature points
- data sets
- repetitive structures
- salient features
- x ray images
- range data
- computer vision