This repository provides an official implementation of High-Resolution Networks (HRNet) combined with Object-Contextual Representations (OCR), rephrased as the Segmentation Transformer. The project targets semantic segmentation, a fundamental computer vision task that assigns class labels to each pixel in an image. It includes models trained on standard benchmarks including Cityscapes, PASCAL-Context, LIP, and ADE20K, and has been integrated into the MMSegmentation framework.