FISBe: A real-world benchmark dataset for instance segmentation of long-range thin filamentous structures

arXiv:2404.00130v1 Announce Type: cross
Abstract: Instance segmentation of neurons in volumetric light microscopy images of nervous systems enables groundbreaking research in neuroscience by facilitating joint functional and morphological analyses of neural circuits at cellular resolution. Yet said multi-neuron light microscopy data exhibits extremely challenging properties for the task of instance segmentation: Individual neurons have long-ranging, thin filamentous and widely branching morphologies, multiple neurons are tightly inter-weaved, and partial volume effects, uneven illumination and noise inherent to light microscopy severely impede …

abstract arxiv benchmark cellular circuits cs.cv cs.lg data dataset functional groundbreaking images instance light microscopy neuron neurons neuroscience research resolution said segmentation systems type world