ShapeClipper: Scalable 3D Shape Learning from Single-View Images via Geometric and CLIP-based Consistency.
Zixuan HuangVarun JampaniAnh ThaiYuanzhen LiStefan StojanovJames M. RehgPublished in: CoRR (2023)
Keyphrases
- single view
- multi view
- three dimensional
- learning algorithm
- polyhedral objects
- learning process
- image data
- perspective projection
- vanishing points
- repetitive structures
- geometric structure
- multiple views
- co training
- input image
- geometric information
- d objects
- active learning
- video clips
- surface reconstruction
- single image
- supervised learning
- semi supervised
- surface geometry
- man made objects