U-Net: Convolutional Networks for Biomedical Image Segmentation
Citations
13,468 citations
11,127 citations
10,120 citations
9,803 citations
Cites background from "U-Net: Convolutional Networks for B..."
...es features for fusion and captures context with global pooling. The “deconvolutional network” approach of [42] restores resolution by proposals, stacks of learned deconvolution, and unpooling. U-Net [43] combines skip layers and learned deconvolution for pixel labeling of microscopy images. The dilation architecture of [44] makes 3 thorough use of dilated convolution for pixel-precise output without ...
[...]
8,730 citations
Cites background or methods from "U-Net: Convolutional Networks for B..."
...An example is the lastic deformations that were applied in the original U-Net paper Ronneberger et al., 2015 )....
[...]
...Ronneberger et al. (2015) took the idea of the fCNN one step further and proposed the U-net architecture, comprising a ‘regu- lar’ fCNN followed by an upsampling part where ‘up’-convolutions are used to increase the image size, coined contractive and expan- sive paths....
[...]
...The most well-known, in medical image analysis, of these ovel CNN architectures is U-net, published by Ronneberger et al. 2015) ( Section 2.4.3 )....
[...]
...The most well-known of these novel CNN architectures is U-net, published by Ronneberger et al. (2015) (section 2....
[...]
References
12,531 citations
11,866 citations
11,732 citations
10,161 citations
"U-Net: Convolutional Networks for B..." refers methods in this paper
...We provide the full Caffe[6]-based implementation and the trained networks4....
[...]
...We provide the full Caffe[6]-based implementation and the trained networks(4)....
[...]
...The input images and their corresponding segmentation maps are used to train the network with the stochastic gradient descent implementation of Caffe [6]....
[...]
9,803 citations
"U-Net: Convolutional Networks for B..." refers background or methods in this paper
...In this paper, we build upon a more elegant architecture, the so-called “fully convolutional network” [9]....
[...]
...The main idea in [9] is to supplement a usual contracting network by successive layers, where pooling operators are replaced by upsampling operators....
[...]