Initial Images: Using Image Prompts to Improve Subject Representation in Multimodal AI Generated Art.
Han QiaoVivian LiuLydia B. ChiltonPublished in: Creativity & Cognition (2022)
Keyphrases
- input image
- image data
- image features
- image collections
- test images
- image retrieval
- image analysis
- original images
- image description
- image set
- high contrast
- image classification
- image pixels
- image representation
- image regions
- pixel values
- feature representation
- reference images
- fundamental problem in computer vision
- illumination conditions
- segmentation method
- image processing algorithms
- intensity values
- grey level
- edge detection
- image noise
- region of interest
- normalized correlation
- multimodal image registration
- imaging process
- raw image
- geometrical information
- lighting conditions
- web images
- image database
- synthesized images
- normalized gradient
- single image
- image matching
- feature points
- image content
- segmented images
- segmentation algorithm
- sample images
- contrast enhancement
- visual patterns
- curvelet transform
- relative position
- geometric shapes
- gray value
- object recognition
- image structure
- spatial information
- secret image
- digital imaging
- bounding box
- set of training images
- image segmentation
- multiscale
- textual descriptions
- bag of features
- visual features
- image registration
- color histogram
- face recognition
- gray level
- keypoints
- natural images
- gaussian noise
- image annotation
- object shapes
- super resolution
- visual data
- feature descriptors