CLIP-GEN: Language-Free Training of a Text-to-Image Generator with CLIP.
Zihao WangWei LiuQian HeXinglong WuZili YiPublished in: CoRR (2022)
Keyphrases
- input image
- image data
- image content
- multiscale
- single image
- image classification
- image features
- image segmentation
- information retrieval
- edge detection
- text mining
- textual information
- image analysis
- image retrieval
- training examples
- image collections
- computational linguistics
- image pixels
- text information
- region of interest
- test images
- image regions
- hough transform
- segmentation method
- similarity measure
- text documents
- spatial information
- key frames
- video clips
- programming language
- complex background
- low level
- text lines
- english text
- text to speech synthesis