EATEN: Entity-aware Attention for Single Shot Visual Text Extraction.
He GuoXiameng QinJiaming LiuJunyu HanJingtuo LiuErrui DingPublished in: CoRR (2019)
Keyphrases
- text extraction
- single shot
- visual features
- complex background
- text processing
- visual information
- visual attention
- natural scenes
- light field
- viewpoint
- text information
- document classification
- spatial resolution
- image reconstruction
- text recognition
- structured light
- human vision
- face detection
- text segmentation
- motion estimation
- high quality