AI RESEARCH
NeuroSeg Meets DINOv3: Transferring 2D Self-Supervised Visual Priors to 3D Neuron Segmentation via DINOv3 Initialization
arXiv CS.CV
•
ArXi:2603.23104v1 Announce Type: new 2D visual foundation models, such as DINOv3, a self-supervised model trained on large-scale natural images, have nstrated strong zero-shot generalization, capturing both rich global context and fine-grained structural cues. However, an analogous 3D foundation model for downstream volumetric neuroimaging remains lacking, largely due to the challenges of 3D image acquisition and the scarcity of high-quality annotations.