GRiT: A Generative Region-to-text Transformer for Object Understanding.
Jialian WuJianfeng WangZhengyuan YangZhe GanZicheng LiuJunsong YuanLijuan WangPublished in: CoRR (2022)
Keyphrases
- image regions
- region of interest
- text retrieval
- d objects
- fuzzy logic
- target object
- generative model
- complex objects
- object regions
- spatial regions
- spatial location
- deformable objects
- object segmentation
- spatial relationships
- text documents
- unsupervised learning
- text mining
- neural network
- power system
- object model
- multiple objects
- deeper understanding
- moving objects
- information retrieval
- spatial extent
- database