ContourRend: a segmentation method for improving contours by rendering

Junwen Chen, Yi Lu, Yaran Chen*, Dongbin Zhao, Zhonghua Pang

*Corresponding author for this work

Research output: Chapter in Book or Report/Conference proceedingConference Proceedingpeer-review

1 Citation (Scopus)

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 contour 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%.

Original languageEnglish
Title of host publicationAdvances in Neural Networks – ISNN 2020 - 17th International Symposium on Neural Networks, ISNN 2020, Proceedings
EditorsMin Han, Sitian Qin, Nian Zhang
PublisherSpringer Science and Business Media Deutschland GmbH
Pages251-260
Number of pages10
ISBN (Print)9783030642204
DOIs
Publication statusPublished - 2020
Event17th International Symposium on Neural Networks, ISNN 2020 - Cairo, Egypt
Duration: 4 Dec 20206 Dec 2020

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume12557 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference17th International Symposium on Neural Networks, ISNN 2020
Country/TerritoryEgypt
CityCairo
Period4/12/206/12/20

Keywords

  • Contour renderer
  • Convolution neural networks
  • Graph convolutional network
  • Image segmentation

Fingerprint

Dive into the research topics of 'ContourRend: a segmentation method for improving contours by rendering'. Together they form a unique fingerprint.

Cite this