前言

要道尾聲了，Unet的結論對於這們短的論文來說，其實算是不太重要的(看了也是)。

Conclusion

The u-net architecture achieves very good performance on very different biomedical segmentation applications.

先商業自吹一下。

Thanks to data augmentation with elastic deformations, it only needs very few annotated images and has a very reasonable training time of only 10 hours on a NVidia Titan GPU (6 GB).

可以在GPU訓練 10 小時。

We provide the full Caffe[6]-based implementation and the trained network. We are sure that the u-net architecture can be applied easily to many more tasks.

提供了基於 caffe 實踐的開源程式碼，可以輕鬆實現。

Concusion

這是一個如果我寫出來會避不了業的結論。但是對方畢竟是很強的研究團隊，在短短12頁裡面受益良多，接下來就是實現部分，明天見。

Reference

[0] U-net