A U-Net++ With Pre-Trained EfficientNet Backbone for Segmentation of Diseases and Artifacts in Endoscopy Images and Videos.
Lê Duy HuynhNicolas BoutryPublished in: EndoCV@ISBI (2020)
Keyphrases
- pre trained
- segmentation algorithm
- segmentation method
- video images
- input image
- training data
- pixel level
- three dimensional
- computer vision
- image registration
- image retrieval
- image features
- image matching
- video frames
- image segmentation
- small number
- neural network
- feature extraction
- training examples
- lighting conditions