History

Junjun2016 5dacca3ea8 Add 4 retinal vessel segmentation benchmark (#315 ) * add 4 retinal vessel segmentation configs of UNet * fix flip augmentation * add unet benchmark on 4 medical datasets * fix hrf bug		2020-12-23 23:58:09 -08:00
..
README.md	Add 4 retinal vessel segmentation benchmark (#315 )	2020-12-23 23:58:09 -08:00
unet_s5-d16_64x64_40k_drive.py	Add 4 retinal vessel segmentation benchmark (#315 )	2020-12-23 23:58:09 -08:00
unet_s5-d16_128x128_40k_chase_db1.py	Add 4 retinal vessel segmentation benchmark (#315 )	2020-12-23 23:58:09 -08:00
unet_s5-d16_128x128_40k_stare.py	Add 4 retinal vessel segmentation benchmark (#315 )	2020-12-23 23:58:09 -08:00
unet_s5-d16_256x256_40k_hrf.py	Add 4 retinal vessel segmentation benchmark (#315 )	2020-12-23 23:58:09 -08:00

README.md

U-Net: Convolutional Networks for Biomedical Image Segmentation

Introduction

@inproceedings{ronneberger2015u,
  title={U-net: Convolutional networks for biomedical image segmentation},
  author={Ronneberger, Olaf and Fischer, Philipp and Brox, Thomas},
  booktitle={International Conference on Medical image computing and computer-assisted intervention},
  pages={234--241},
  year={2015},
  organization={Springer}
}

Results and models

Backbone	Head	Dataset	Image Size	Crop Size	Stride	Lr schd	Mem (GB)	Inf time (fps)	Dice	download
UNet-S5-D16	FCN	DRIVE	584x565	64x64	42x42	40000	0.680	-	78.67	model \| log
UNet-S5-D16	FCN	STARE	605x700	128x128	85x85	40000	0.968	-	81.02	model \| log
UNet-S5-D16	FCN	CHASE_DB1	960x999	128x128	85x85	40000	0.968	-	80.24	model \| log
UNet-S5-D16	FCN	HRF	2336x3504	256x256	170x170	40000	2.525	-	79.45	model \| log