Self-supervised Scene Text Segmentation with Object-centric Layered Representations Augmented by Text Regions.
Yibo WangYunhu YeYuanpeng MaoYanwei YuYuanping SongPublished in: CoRR (2023)
Keyphrases
- text segmentation
- scene images
- text regions
- bounding box
- input image
- multiple objects
- moving objects
- target object
- video images
- document images
- character recognition
- optical character recognition
- image regions
- connected components
- video frames
- d objects
- complex scenes
- multiscale
- object tracking
- text information
- complex background
- lighting conditions
- object classes
- eye movements
- binary images
- image features
- video sequences