DSEG-LIME - Improving Image Explanation by Hierarchical Data-Driven Segmentation.
Patrick KnabSascha MartonChristian BarteltPublished in: CoRR (2024)
Keyphrases
- data driven
- segmentation algorithm
- segmentation method
- image segmentation
- multiscale
- hierarchical segmentation
- image analysis
- gray level images
- image regions
- grey level
- hierarchical representation
- input image
- adaptive thresholding
- optimal segmentation
- single image
- image data
- image retrieval
- image content
- segmented images
- test images
- energy function
- edge detection
- microscopic images
- image classification
- joint segmentation
- low level image segmentation
- pixel level
- image features
- berkeley segmentation dataset
- region of interest
- segmentation errors
- image representation
- pixel wise
- perceptual grouping
- segmentation accuracy
- energy functional
- images of natural scenes
- colour images
- region segmentation
- coarse to fine
- image segments
- brain mr images
- watershed algorithm
- foreground background separation
- unsupervised image segmentation
- binary partition tree
- watershed segmentation
- image segmentation algorithm
- medical images
- gray value
- homogeneous regions
- level set
- bounding box
- pixel values
- region growing
- microscope images
- image segmentation algorithms
- optical flow
- watershed transformation
- color features
- circular hough transform
- image pixels