Optimized latent-code selection for explainable conditional text-to-image GANs.
Zhenxing ZhangLambert SchomakerPublished in: IJCNN (2022)
Keyphrases
- multiscale
- input image
- image data
- single image
- image retrieval
- image representation
- image classification
- image features
- image content
- image analysis
- pixel values
- web images
- segmentation method
- region of interest
- template matching
- latent variables
- image pixels
- image collections
- low level
- image segmentation
- image regions
- test images
- information retrieval
- feature points
- similarity measure
- spatial information
- hough transform
- text retrieval
- image quality
- random fields
- edge detection
- bounding box
- textual information
- binary codes
- text information