A New Dataset Based on Images Taken by Blind People for Testing the Robustness of Image Classification Models Trained for ImageNet Categories.
Reza Akbarian BafghiDanna GurariPublished in: CVPR (2023)
Keyphrases
- image collections
- million images
- classification models
- image dataset
- input image
- image data
- image set
- image features
- test images
- image content
- image retrieval
- image regions
- image classification
- image search
- sample images
- image database
- pre trained
- web images
- image annotation
- photo collections
- segmentation method
- street view
- pixel values
- object categories
- category specific
- image matching
- edge detection
- visual features
- image noise
- feature points
- image representation
- feature set
- visual attributes
- segmentation algorithm
- training data
- scene categories
- secret image
- object recognition
- visual similarity
- decision trees
- neural network
- vector quantized
- set of training images
- high resolution
- training set
- feature selection
- computer vision