Does an ensemble of GANs lead to better performance when training segmentation networks with synthetic images?
Måns LarssonMuhammad Usman AkbarAnders EklundPublished in: CoRR (2022)
Keyphrases
- training set
- level set
- medical images
- neural network ensemble
- segmentation algorithm
- multiscale
- recurrent networks
- segmentation method
- image segmentation
- training data
- edge detection
- training samples
- training examples
- segmentation accuracy
- training process
- shape prior
- neural network
- image analysis
- social networks
- test set
- fully automatic
- network structure
- region growing
- ensemble learning
- machine learning
- learning algorithm
- fully unsupervised
- random forests
- decision trees
- bounding box
- object segmentation
- multi class
- energy function