Home
- What is DenoiSeg?
- How to use DenoiSeg?
- Qualitative Results
- Quantitative Performance Comparisons
- Contact us
DenoiSeg is a deep learning based segmentation method which can leverage unsupervised denoising with Noise2Void for cell/nuclei segmentation using only limited amount of ground truth segmentation annotations. The key contribution is to train a network for performing denoising and segmentation jointly. The reason for the success of DenoiSeg 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. 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. For more details, please see the paper.
DenoiSeg is available as a pip installable package. Just follow these simple steps here to get DenoiSeg running on your machine. We also provide example Jupyter notebooks to try DenoiSeg for some publicly available datasets from our publication.
We show results on three diverse datasets: DSB 2018, a developing Fly Wing and a Mouse Nuclei dataset. For each dataset, we add Gaussian noise with mean 0 and standard deviation 10 and 20. The dataset names are accordingly extended by n0, n10 and n20 to indicate the amount of additional noise. We compare against two baselines: (i) DenoiSeg trained purely for segmentation (referred to as Baseline), and a sequential scheme based on Prakash et. al that first trains a denoiser and then the aforementioned baseline (referred to as Sequential).
DenoiSeg results using only only 10, 2 and 2 ground truth segmentation annotations for DSB, Fly Wing and Mouse Nuclei dataset respectively:
DenoiSeg results using only only 38, 7 and 5 ground truth segmentation annotations for DSB, Fly Wing and Mouse Nuclei dataset respectively:
DenoiSeg results using only only 304, 29 and 18 ground truth segmentation annotations for DSB, Fly Wing and Mouse Nuclei dataset respectively:
We compare the results of DenoiSeg with the above mentioned baselines in terms of AP score and SEG score. For each dataset and each noise level, the tables present DenoiSeg setups with different values of hyperparameter Alpha. Additionally we also compare to results that use (the a priori unknown) best Alpha. Please refer to the paper to see how the best Alpha is found. The figures corresponding to the tables below can be found in the paper.
Result tables of Flywing dataset.
Result tables of Mouse dataset.
Additive Gaussian Noise Augmentation.
Performance Comparison with Other Semi-Supervised Segmentation Routines
If you have questions about our methods or have issues with our code, please open an issue and we will get back to you as soon as possible. Enjoy segmenting and denoising your data with DenoiSeg!