CATNet: Scene Text Recognition Guided by Concatenating Augmented Text Features.
Ziyin ZhangLemeng PanLin DuQingrui LiNing LuPublished in: ICDAR (1) (2021)
Keyphrases
- additional features
- scene text recognition
- automatically extracted
- feature vectors
- keywords
- neural network
- feature set
- classification accuracy
- structural features
- text retrieval
- structural information
- image classification
- co occurrence
- low level
- similarity measure
- feature extraction
- decision trees
- computer vision
- information retrieval