Iterative Interaction Training for Segmentation Editing Networks
By Gustav Bredell, Christine Tanner, Ender Konukoglu. 2018
36 citations recorded by the source corpus.
Automatic segmentation has great potential to facilitate morphological measurements while simultaneously increasing efficiency. Nevertheless often users want to edit the segmentation to their own needs and will need different tools for this. There has been methods developed to edit segmentations of automatic methods based on the user input, primarily for binary segmentations. Here however, we present an unique training strategy for convolutional neural networks (CNNs) trained on top of an automatic method to enable interactive segmentation editing that is not limited to binary segmentation. By utilizing a robot-user during training, we closely mimic realistic use cases to achieve optimal editing performance. In addition, we show that an increase of the iterative interactions during the training process up to ten improves the segmentation editing performance substantially. Furthermore, we compare our segmentation editing CNN (interCNN) to state-of-the-art interactive segmentation algorithms and show a superior or on par performance.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
In Computer Vision and Pattern Recognition
- Brian L. Price
- Scott D. Cohen
- Ning Xu
- Franccois Piti'e
- Marco Forte
- Luc Van Gool
- Jiashi Feng
- Alan L. Yuille
Citations in the corpus
- Getting to 99% Accuracy in Interactive Segmentation (2020, methodology): “Bredell et al. [18] is to feed back the mask alongside the image and the clicks (early fusion) in order to retain some state information.”
- A Survey on Active Learning and Human-in-the-Loop Deep Learning for Medical Image Analysis (2019, unknown): “Bredell et al. (2018) propose an alternative to BIFSeg in which two networks are trained, one to perform an initial segmentation (they use a CNN but this initial segmentation could be performed with any existing algorithm) and a second network they call interCNN that takes as input the image, some…”