Many real-world vision problems suffer from inherent ambiguities. In clinical applications for example, it might not be clear from a CT scan alone which particular region is cancer tissue. Therefore a group of graders typically produces a set of diverse but plausible segmentations. We consider the task of learning a distribution over segmentations given an input. To this end we propose a generative segmentation model based on a combination of a U-Net with a conditional variational autoencoder that is capable of efficiently producing an unlimited number of plausible hypotheses. We show on a lung abnormalities segmentation task and on a Cityscapes segmentation task that our model reproduces the possible segmentation variants as well as the frequencies with which they occur, doing so significantly better than published approaches. These models could have a high impact in real-world applications, such as being used as clinical decision-making algorithms accounting for multiple plausible semantic segmentation hypotheses to provide possible diagnoses and recommend further actions to resolve the present ambiguities.

Precise image segmentation provides clinical study with instructive information.

Specifically, SegRefiner takes coarse masks as inputs and refines them using a discrete diffusion process.

"freely" into massive pre-training datasets to support large-scale pre-training.

Uncertainty Quantification (UQ) is crucial for reliable image segmentation.

Many real-world vision problems suffer from inherent ambiguities. In clinical applications for example, itmight not be clear from aCT scan alone which particular region is cancer tissue. Therefore a group of graders typically produces a set of diverse but plausible segmentations.