Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: May 8, 2026

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

SCASeg: Strip Cross-Attention for Efficient Semantic Segmentation.

Guoan Xu, Jiaming Chen, Wenfeng Huang

    IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
    |May 6, 2026
    PubMed
    Summary
    This summary is machine-generated.

    Related Concept Videos

    You might also read

    Related Articles

    Articles linked to this work by shared authors, journal, and citation graph.

    Sort by
    Same author

    Daily fall risk assessment for older adults using a balance sensor with machine learning.

    Frontiers in public health·2026
    Same author

    Single-cell analysis of fetal testis reveals dysfunction of human Leydig cells in Klinefelter syndrome.

    The Journal of clinical investigation·2026
    Same author

    Molecular Evolution of Organic Matter Humification Governs Ferrihydrite Transformation: Interfacial Electron Transfer Mechanisms and Carbon Preservation Implications.

    Environmental science & technology·2026
    Same author

    Microscopic Control of Coal Quality and Pore Structure on CH<sub>4</sub> Adsorption of Different-Rank Coals: A Multifractal Perspective.

    Langmuir : the ACS journal of surfaces and colloids·2026
    Same author

    Half-Sandwich Ir(III) and Ru(II) Complexes Featuring a Coumarin-Based Chelating Core for Mitochondrial-Targeted Anticancer Activity.

    Inorganic chemistry·2026
    Same author

    Recombinant humanized type III collagen improves ovarian function via ITGA2-mediated mitochondrial function restoration in granulosa cells and extracellular matrix remodeling.

    Regenerative biomaterials·2026

    This study introduces Strip Cross-Attention (SCASeg), a novel decoder for efficient semantic segmentation using Vision Transformers (ViT). SCASeg enhances feature interaction and speeds up inference, outperforming existing methods on benchmark datasets.

    Area of Science:

    • Computer Vision
    • Deep Learning
    • Image Segmentation

    Background:

    • Vision Transformers (ViT) are successful general-purpose visual encoders.
    • ViT backbones require specialized decoders for tasks like semantic segmentation.
    • Existing decoders may not fully leverage ViT capabilities for segmentation.

    Purpose of the Study:

    • To design an efficient and effective decoder head for semantic segmentation using ViT.
    • To improve feature interaction and computational efficiency in ViT-based segmentation models.
    • To introduce a novel architecture, Strip Cross-Attention (SCASeg), for semantic segmentation.

    Main Methods:

    • Proposed Strip Cross-Attention (SCASeg) decoder head.
    • Utilized lateral connections with encoder features as Queries.

    More Related Videos

    Automated Analysis of C. elegans Fluorescence Images using SegElegans
    06:27

    Automated Analysis of C. elegans Fluorescence Images using SegElegans

    Published on: October 10, 2025

    Related Experiment Videos

    Last Updated: May 8, 2026

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
    04:48

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

    Published on: July 5, 2024

    Automated Analysis of C. elegans Fluorescence Images using SegElegans
    06:27

    Automated Analysis of C. elegans Fluorescence Images using SegElegans

    Published on: October 10, 2025

  • Introduced Cross-Layer Block (CLB) for unified Keys and Values representation.
  • Incorporated convolution for local context and compressed channels for efficiency.

    • Main Results:

    • SCASeg demonstrates competitive performance across various setups.
    • Outperformed leading segmentation architectures on ADE20K, Cityscapes, COCO-Stuff 164k, and Pascal VOC2012.
    • Achieved improved computational efficiency, reduced memory usage, and increased inference speed.

    Conclusions:

    • SCASeg is an adaptable and efficient decoder for semantic segmentation.
    • The proposed methods effectively capture global and local context dependencies.
    • SCASeg offers a promising alternative to conventional decoders for ViT-based segmentation.