Perceptual Quality-preserving Black-Box Attack against Deep Learning Image Classifiers.
Diego GragnanielloFrancesco MarraGiovanni PoggiLuisa VerdolivaPublished in: CoRR (2019)
Keyphrases
- black box
- perceptual quality
- deep learning
- image data
- image quality
- input image
- image quality assessment
- visual quality
- image features
- image representation
- image retrieval
- human perception
- test images
- image content
- single image
- texture synthesis
- image segmentation
- image classification
- human visual system
- structural similarity
- image regions
- video quality
- quality measures
- high resolution
- multiscale
- feature selection
- principal component analysis
- low level
- similarity measure