ViewCo: Discovering Text-Supervised Segmentation Masks via Multi-View Semantic Consistency.
Pengzhen RenChanglin LiHang XuYi ZhuGuangrun WangJianzhuang LiuXiaojun ChangXiaodan LiangPublished in: CoRR (2023)
Keyphrases
- multi view
- semi supervised
- single view
- semantic information
- multiple views
- three dimensional
- depth map
- camera calibration
- d objects
- multiple cameras
- surface reconstruction
- multi view reconstruction
- learning algorithm
- text mining
- multi view face detection
- multi view images
- co training
- scene reconstruction
- range images
- view synthesis
- light field
- multiple viewpoints
- multi view stereo
- supervised learning
- unsupervised learning
- visual hull
- information extraction
- depth information
- pairwise
- depth video