English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Several imaging applications (vessels, retina, plant roots, road networksfrom satellites) require the accurate segmentation of thin structures forsubsequent analysis. Discontinuities (gaps) in the extracted foreground mayhinder down-stream image-based analysis of biomarkers, organ structure andtopology. In this paper, we propose a general post-processing technique torecover such gaps in large-scale segmentation masks. We cast this problem as ablind inpainting task, where the regions of missing lines in the segmentationmasks are not known to the algorithm, which we solve with an adversariallytrained neural network. One challenge of using large images is the memorycapacity of current GPUs. The typical approach of dividing a large image intosmaller patches to train the network does not guarantee global coherence of thereconstructed image that preserves structure and topology. We use adversarialtraining and reinforcement learning (Policy Gradient) to endow the model withboth global context and local details. We evaluate our method in severaldatasets in medical imaging, plant science, and remote sensing. Our experimentsdemonstrate that our model produces the most realistic and complete inpaintedresults, outperforming other approaches. In a dedicated study on plant roots wefind that our approach is also comparable to human performance. Implementationavailable at \url{https://github.com/Hao/Thin-Structure-Inpainting}.

Blind Inpainting of Large-scale Masks of Thin Structures with Adversarial and Reinforcement Learning