How Does Frequency Bias Affect the Robustness of Neural Image Classifiers against Common Corruption and Adversarial Perturbations?
Alvin ChanYew-Soon OngClement TanPublished in: CoRR (2022)
Keyphrases
- image data
- input image
- multiscale
- image features
- image content
- single image
- image classification
- neural network
- image analysis
- segmentation method
- low frequency
- template matching
- image representation
- feature points
- feature selection
- test images
- image matching
- similarity measure
- image segmentation
- decision trees
- image retrieval
- image collections
- image set
- high frequency
- training data
- multi agent
- network architecture
- image pixels
- high resolution
- image transformations
- classifier training
- edge detection
- training set
- object recognition
- support vector
- learning algorithm