← All works
Research paper

U-Net: Convolutional Networks for Biomedical Image Segmentation

At a glance

U-Net outlines cells and other structures in biomedical images.

In one paragraph

U-Net outlines cells and other structures in biomedical images by combining broad visual context with fine detail. Its design links earlier and later processing stages to preserve precise locations. The paper showed strong results even with relatively few labeled training images.

Authors 3

  1. Olaf Ronneberger
  2. Philipp Fischer
  3. Thomas Brox