ContourRend: A Segmentation Method for Improving Contours by Rendering

TL;DR

ContourRend is a novel segmentation approach that refines object contours by using a contour renderer, significantly improving edge clarity and accuracy over existing methods.

Contribution

The paper introduces ContourRend, a contour rendering technique integrated with GCN-based segmentation to enhance contour detail and segmentation quality.

Findings

Achieves 72.41% mean IoU on Cityscapes.

Surpasses Polygon-GCN baseline by 1.22%.

Effectively refines contours with high-resolution rendering.

Abstract

A good object segmentation should contain clear contours and complete regions. However, mask-based segmentation can not handle contour features well on a coarse prediction grid, thus causing problems of blurry edges. While contour-based segmentation provides contours directly, but misses contours' details. In order to obtain fine contours, we propose a segmentation method named ContourRend which adopts a contour renderer to refine segmentation contours. And we implement our method on a segmentation model based on graph convolutional network (GCN). For the single object segmentation task on cityscapes dataset, the GCN-based segmentation con-tour is used to generate a contour of a single object, then our contour renderer focuses on the pixels around the contour and predicts the category at high resolution. By rendering the contour result, our method reaches 72.41% mean intersection over union (IoU) and surpasses baseline Polygon-GCN by 1.22%.