A Friendly Face: Do Text-to-Image Systems Rely on Stereotypes when the Input is Under-Specified?
Kathleen C. FraserSvetlana KiritchenkoIsar NejadgholiPublished in: CoRR (2023)
Keyphrases
- image data
- input image
- multiscale
- image features
- image content
- image analysis
- image classification
- image representation
- segmentation method
- lighting conditions
- image segmentation
- image retrieval
- image regions
- text information
- template matching
- single image
- illumination conditions
- image matching
- region of interest
- edge detection
- low level
- high resolution
- hand drawn
- information retrieval
- image set
- hough transform
- web images
- complex background
- text input
- pixel values
- image pixels
- textual and visual information
- test images
- spatial information
- segmentation algorithm
- feature points
- input data
- face images
- object recognition
- image sequences
- computer vision