What can we learn from misclassified ImageNet images?
Shixian WenAmanda Sofie RiosKiran LekkalaLaurent IttiPublished in: CoRR (2022)
Keyphrases
- image collections
- image database
- image retrieval
- image data
- three dimensional
- image analysis
- image classification
- image registration
- ground truth
- image features
- object recognition
- input image
- multiple images
- image regions
- rigid body
- test images
- feature points
- computer graphics
- edge detection
- region of interest
- image understanding
- similarity measure
- million images
- image processing
- lighting conditions
- principal components
- illumination conditions
- pixel values
- training data
- small number