Loughborough University
Browse

A multi-strategy contrastive learning framework for weakly supervised semantic segmentation

Download (4.83 MB)
journal contribution
posted on 2023-01-23, 10:25 authored by Colin Yuan, Gerald SchaeferGerald Schaefer, Yu-Kun Lai, Yifan Wang, Xiyao Liu, Lin GuanLin Guan, Hui FangHui Fang
Weakly supervised semantic segmentation (WSSS) has gained significant popularity as it relies only on weak labels such as image level annotations rather than the pixel level annotations required by supervised semantic segmentation (SSS) methods. Despite drastically reduced annotation costs, typical feature representations learned from WSSS are only representative of some salient parts of objects and less reliable compared to SSS due to the weak guidance during training. In this paper, we propose a novel Multi-Strategy Contrastive Learning (MuSCLe) framework to obtain enhanced feature representations and improve WSSS performance by exploiting similarity and dissimilarity of contrastive sample pairs at image, region, pixel and object boundary levels. Extensive experiments demonstrate the effectiveness of our method and show that MuSCLe outperforms current state-of-the-art methods on the widely used PASCAL VOC 2012 dataset.

History

School

  • Science

Department

  • Computer Science

Published in

Pattern Recognition

Volume

137

Issue

2023

Publisher

Elsevier

Version

  • VoR (Version of Record)

Rights holder

© The Author(s)

Publisher statement

This is an Open Access Article. It is published by Elsevier under the Creative Commons Attribution 4.0 International Licence (CC BY). Full details of this licence are available at: http://creativecommons.org/licenses/by/4.0/

Acceptance date

2022-12-31

Publication date

2023-01-02

Copyright date

2023

ISSN

0031-3203

Language

  • en

Depositor

Dr Hui Fang. Deposit date: 21 January 2023

Article number

109298

Usage metrics

    Loughborough Publications

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC