EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Enhancing Mask Transformer with Auxiliary Convolution Layers for Semantic Segmentation

Zhengyu Xia iD, Joohee Kim iD

Sensors · 2023

DOI: 10.3390/s23020581 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, 2023. Verfasst von Zhengyu Xia und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electrical and Computer Engineering, Illinois Institute of Technology, Chicago, IL 60616, USA. Hauptthemen laut MEDLINE: Semantics, Electric Power Supplies. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Trade, Industry and Energy, Technology Innovation Program of the Ministry of Trade, Industry & Energy (MOTIE, Republic of Korea).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Transformer-based semantic segmentation methods have achieved excellent performance in recent years. Mask2Former is one of the well-known transformer-based methods which unifies common image segmentation into a universal model. However, it performs relatively poorly in obtaining local features and segmenting small objects due to relying heavily on transformers. To this end, we propose a simple yet effective architecture that introduces auxiliary branches to Mask2Former during training to capture dense local features on the encoder side. The obtained features help improve the performance of learning local information and segmenting small objects. Since the proposed auxiliary convolution layers are required only for training and can be removed during inference, the performance gain can be obtained without additional computation at inference. Experimental results show that our model can achieve state-of-the-art performance (57.6% mIoU) on the ADE20K and (84.8% mIoU) on the Cityscapes datasets.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

SemanticsElectric Power SuppliesLearningImage Processing, Computer-Assisted

Stichwörter der Autor:innen

convolutional neural networksdeep learningimage segmentationsemantic segmentationtransformer

Autor:innen und Institutionen

  • Department of Electrical and Computer Engineering, Illinois Institute of Technology, Chicago, IL 60616, USA.

ORCID-Profile: Zhengyu Xia, Joohee Kim

Förderung

  • Ministry of Trade, Industry and Energy
  • Technology Innovation Program of the Ministry of Trade, Industry & Energy (MOTIE, Republic of Korea)

Zitieren

APA

Xia, Z., & Kim, J. (2023). Enhancing Mask Transformer with Auxiliary Convolution Layers for Semantic Segmentation. Sensors. https://doi.org/10.3390/s23020581

Vancouver

Xia Z, Kim J. Enhancing Mask Transformer with Auxiliary Convolution Layers for Semantic Segmentation. Sensors. 2023. doi:10.3390/s23020581

BibTeX
@article{Xia2023,
  author = {Xia, Zhengyu and Kim, Joohee},
  title = {Enhancing Mask Transformer with Auxiliary Convolution Layers for Semantic Segmentation},
  journal = {Sensors},
  year = {2023},
  doi = {10.3390/s23020581}
}
Kostenlos starten →