Weakly Supervised Scene Text Generation for Low-resource Languages.
Yangchen XieXinyuan ChenHongjian ZhanPalaiahankote ShivakumaraBing YinCong LiuYue LuPublished in: CoRR (2023)
Keyphrases
- weakly supervised
- text generation
- natural language generation
- superpixels
- object detectors
- topic models
- object class
- input image
- multiple images
- natural language
- named entities
- three dimensional
- relation extraction
- video sequences
- single image
- object recognition
- multiple objects
- object detection
- semi supervised
- image sequences
- viewpoint
- theorem prover
- moving objects
- domain specific
- dialogue system
- multiscale
- automatic extraction
- co occurrence
- multiple views
- image regions
- machine learning
- natural images