Perceptual quality-preserving black-box attack against deep learning image classifiers.
Diego GragnanielloFrancesco MarraLuisa VerdolivaGiovanni PoggiPublished in: Pattern Recognit. Lett. (2021)
Keyphrases
- black box
- perceptual quality
- deep learning
- image quality assessment
- visual quality
- image quality
- image data
- human perception
- input image
- image content
- image representation
- single image
- image classification
- image segmentation
- low level
- multiscale
- image features
- texture synthesis
- human visual system
- image retrieval
- structural similarity
- image regions
- test images
- image processing
- decision trees
- high level
- quality measures
- training data
- high quality
- test cases
- test set
- feature space
- machine learning
- high resolution