Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

DenoiSeg: Joint Denoising and Segmentation

About

Microscopy image analysis often requires the segmentation of objects, but training data for this task is typically scarce and hard to obtain. Here we propose DenoiSeg, a new method that can be trained end-to-end on only a few annotated ground truth segmentations. We achieve this by extending Noise2Void, a self-supervised denoising scheme that can be trained on noisy images alone, to also predict dense 3-class segmentations. The reason for the success of our method is that segmentation can profit from denoising, especially when performed jointly within the same network. The network becomes a denoising expert by seeing all available raw data, while co-learning to segment, even if only a few segmentation labels are available. This hypothesis is additionally fueled by our observation that the best segmentation results on high quality (very low noise) raw data are obtained when moderate amounts of synthetic noise are added. This renders the denoising-task non-trivial and unleashes the desired co-learning effect. We believe that DenoiSeg offers a viable way to circumvent the tremendous hunger for high quality training data and effectively enables few-shot learning of dense segmentations.

Tim-Oliver Buchholz, Mangal Prakash, Alexander Krull, Florian Jug• 2020

Related benchmarks

TaskDatasetResultRank
Medical Image SegmentationCOVID-CT
Dice (%)59.29
45
Medical Image SegmentationTN3K
DSC82.6
34
Medical Image SegmentationACDC
DSC64.23
22
Medical Image RestorationACDC
PSNR22.35
18
Medical Image SegmentationHCC-TACE-Seg
Dice Score60.01
15
Medical Image SegmentationACDC
Average mIoU0.8327
9
Medical Image SegmentationKiTS19
Average mIoU67.8
9
Medical Image SegmentationAverage
Dice Coefficient0.5621
9
Medical Image RestorationCOVID19CTscans
PSNR12.96
7
Medical Image RestorationHCC-TACE-Seg
PSNR24.25
7
Showing 10 of 12 rows

Other info

Follow for update