Filter
Associated Lab
- Aso Lab (2) Apply Aso Lab filter
- Card Lab (1) Apply Card Lab filter
- Fitzgerald Lab (1) Apply Fitzgerald Lab filter
- Remove Funke Lab filter Funke Lab
- Hess Lab (2) Apply Hess Lab filter
- Keller Lab (1) Apply Keller Lab filter
- Lippincott-Schwartz Lab (1) Apply Lippincott-Schwartz Lab filter
- Rubin Lab (1) Apply Rubin Lab filter
- Saalfeld Lab (3) Apply Saalfeld Lab filter
- Scheffer Lab (1) Apply Scheffer Lab filter
- Turaga Lab (1) Apply Turaga Lab filter
- Turner Lab (1) Apply Turner Lab filter
Associated Project Team
Associated Support Team
Publication Date
- December 2023 (1) Apply December 2023 filter
- October 2023 (2) Apply October 2023 filter
- July 2023 (2) Apply July 2023 filter
- June 2023 (2) Apply June 2023 filter
- May 2023 (1) Apply May 2023 filter
- February 2023 (1) Apply February 2023 filter
- January 2023 (3) Apply January 2023 filter
- Remove 2023 filter 2023
12 Janelia Publications
Showing 11-12 of 12 resultsSegmentation of objects in microscopy images is required for many biomedical applications. We introduce object-centric embeddings (OCEs), which embed image patches such that the spatial offsets between patches cropped from the same object are preserved. Those learnt embeddings can be used to delineate individual objects and thus obtain instance segmentations. Here, we show theoretically that, under assumptions commonly found in microscopy images, OCEs can be learnt through a self-supervised task that predicts the spatial offset between image patches. Together, this forms an unsupervised cell instance segmentation method which we evaluate on nine diverse large-scale microscopy datasets. Segmentations obtained with our method lead to substantially improved results, compared to state-of-the-art baselines on six out of nine datasets, and perform on par on the remaining three datasets. If ground-truth annotations are available, our method serves as an excellent starting point for supervised training, reducing the required amount of ground-truth needed by one order of magnitude, thus substantially increasing the practical applicability of our method. Source code is available at github.com/funkelab/cellulus.