Paper ID: 2203.11692

Panoptic segmentation with highly imbalanced semantic labels

Josef Lorenz Rumberger, Elias Baumann, Peter Hirsch, Andrew Janowczyk, Inti Zlobec, Dagmar Kainmueller

We describe here the panoptic segmentation method we devised for our participation in the CoNIC: Colon Nuclei Identification and Counting Challenge at ISBI 2022. Key features of our method are a weighted loss specifically engineered for semantic segmentation of highly imbalanced cell types, and a state-of-the art nuclei instance segmentation model, which we combine in a Hovernet-like architecture.

Submitted: Mar 3, 2022