LIIF-GAN: Learning Representation With Local Implicit Image Function and GAN for Realistic Images on a Continuous Scale.
Jun Seok KangSang Chul AhnPublished in: BMVC (2022)
Keyphrases
- input image
- image data
- image features
- image collections
- test images
- image classification
- image retrieval
- image analysis
- high contrast
- image regions
- segmentation method
- original images
- image representation
- feature representation
- image description
- region of interest
- image pixels
- edge detection
- feature points
- perceptual information
- image structure
- imaging process
- single scale
- image database
- image content
- image segmentation
- linear predictors
- image set
- web images
- keypoints
- scale space theory
- multiscale
- digital imaging
- intensity values
- multiple scales
- pixel values
- normalized gradient
- synthesized images
- sample images
- unlabeled images
- band limited
- segmentation algorithm
- invariant descriptors
- lighting conditions
- scale space
- object recognition
- illumination conditions
- object shapes
- gaussian derivatives
- log polar
- fundamental problem in computer vision
- image sequences
- image annotation
- image matching
- image transformations
- single image
- spatial information
- noise free image
- high resolution
- image registration
- image quality
- natural images
- spatial coordinates
- visual features
- visual words
- feature descriptors
- contrast enhancement
- object detectors
- continuous domains
- object shape