HookNet: multi-resolution convolutional neural networks for semantic segmentation in histopathology whole-slide images.
Mart van RijthovenMaschenka BalkenholKarina SilinaJeroen van der LaakFrancesco CiompiPublished in: CoRR (2020)
Keyphrases
- street scenes
- semantic segmentation
- multiresolution
- image data
- input image
- test images
- image database
- image classification
- ground truth
- convolutional neural networks
- image set
- image understanding
- multiple images
- small number
- image features
- superpixels
- pixel level
- label transfer
- image retrieval
- three dimensional
- object recognition
- image annotation
- segmentation method
- region of interest
- bounding box
- image content
- object class
- spatial information
- scene classification
- natural images
- image registration
- viewpoint