SEP: Stage-Enhanced Panoptic Segmentation based on Fully Convolutional Networks

Shucheng Ji, Xiaochen Yuan, Junqi Bao

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


Panoptic segmentation is a critical technology in the field of multimedia, applicable to various domains such as autonomous driving and image recognition. However, due to the enormity and complexity of the task, enhancing the efficiency and accuracy of panoptic segmentation remains a challenge. In this paper, we propose a stage-enhanced panoptic segmentation method which improves the feature extraction network of the backbone, incorporates a stage feature fusion network, and designs a module for adaptive stage feature weight allocation. These enhancements optimize the overall network and enrich the stage features. Experimental results on the publicly available COCO-2017 dataset confirm the performance of Stage-Enhanced Panoptic and demonstrate its superiority compared to other methods.

Original languageEnglish
Title of host publicationFifteenth International Conference on Signal Processing Systems, ICSPS 2023
EditorsZhenkai Zhang, Cheng Li
ISBN (Electronic)9781510675056
Publication statusPublished - 2024
Event15th International Conference on Signal Processing Systems, ICSPS 2023 - Xi'an, China
Duration: 17 Nov 202319 Nov 2023

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X


Conference15th International Conference on Signal Processing Systems, ICSPS 2023


  • Deep Learning
  • Image Segmentation
  • Neural Network
  • Panoptic Segmentation

Cite this