REPROT: Explaining the predictions of complex deep learning architectures for object detection through reducts of an image.
Marilyn BelloGonzalo NápolesLeonardo ConcepciónRafael BelloPablo MesejoOscar CordónPublished in: Inf. Sci. (2024)
Keyphrases
- deep learning
- object detection
- single image
- image classification
- input image
- image content
- image retrieval
- image representation
- image features
- multiscale
- image segmentation
- region of interest
- machine learning
- image regions
- spatial information
- unsupervised learning
- high resolution
- test images
- decision trees
- object recognition
- pairwise
- bounding box