JLInst: Boundary-Mask Joint Learning for Instance Segmentation

Xiaodong Zhao, Junliang Chen, Zepeng Huang, Linlin Shen

Research output: Chapter in Book/Conference proceedingConference contributionpeer-review


Lots of methods have been proposed to improve instance segmentation performance. However, the mask produced by state-of-the-art segmentation networks is still coarse and does not completely align with the whole object instance. Moreover, we find that better object boundary information can help instance segmentation network produce more distinct and clear object masks. Therefore, we present a simple yet effective instance segmentation framework, termed JLInst (Boundary-Mask Joint Learning for Instance Segmentation). Our methods can jointly exploit object boundary and mask semantic information in the instance segmentation network, and generate more precise mask prediction. Besides, we propose the Adaptive Gaussian Weighted Binary Cross-Entropy Loss (GW loss), to focus more on uncertain examples in pixel-level classification. Experiments show that JLInst achieves improved performance (+3.0% AP) than Mask R-CNN on COCO test-dev2017 dataset, and outperforms most recent methods in the fair comparison.

Original languageEnglish
Title of host publicationPattern Recognition and Computer Vision - 6th Chinese Conference, PRCV 2023, Proceedings
EditorsQingshan Liu, Hanzi Wang, Rongrong Ji, Zhanyu Ma, Weishi Zheng, Hongbin Zha, Xilin Chen, Liang Wang
PublisherSpringer Science and Business Media Deutschland GmbH
Number of pages13
ISBN (Print)9789819985548
Publication statusPublished - 2024
Externally publishedYes
Event6th Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2023 - Xiamen, China
Duration: 13 Oct 202315 Oct 2023

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume14436 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference6th Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2023


  • Boundary Information Enhancement
  • Instance Segmentation

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'JLInst: Boundary-Mask Joint Learning for Instance Segmentation'. Together they form a unique fingerprint.

Cite this