Home

The U-Net is one of the most popular neural network architectures for biomedical image segmentation. Its fully convolutional architecture allows overlap-tile-processing for fast and pixel-accurate segmentation of 2D and 3D image data of arbitrary extents. A Megapixel-Image can be processed in less than a second.