Universal Adversarial Perturbations to Understand Robustness of Texture vs. Shape-biased Training.
Kenneth T. CoLuis Muñoz-GonzálezLeslie KanthanBen GlockerEmil C. LupuPublished in: CoRR (2019)
Keyphrases
- pixel intensities
- texture information
- texture features
- multi agent
- color information
- training examples
- supervised learning
- neural network
- feature space
- texture images
- training set
- shape descriptors
- shape from shading
- local binary pattern
- texture classification
- shape model
- shape features
- multiresolution
- texture synthesis
- single image
- multiple scales
- input image