Scene Text Image Super-resolution based on Text-conditional Diffusion Models.
Chihiro NoguchiShun FukudaMasao YamanakaPublished in: WACV (2024)
Keyphrases
- image super resolution
- scene text
- diffusion models
- text detection
- natural scene images
- dictionary learning
- low resolution
- super resolution
- diffusion model
- image registration
- information diffusion
- text regions
- high resolution
- complex background
- sparse representation
- social networks
- connected components
- text mining
- sparse coding
- video frames
- object detection
- text lines
- text information
- maximum a posteriori
- face detection
- probabilistic model
- object recognition