3D Image Annotation

3D Image Annotation is a methodological framework in computer vision that involves assigning semantic labels to three-dimensional data points or structures within volumetric or spatial datasets, distinguishing it from 2D annotation through its explicit modeling of depth, geometry, and spatial relationships in three-dimensional space.

49

Publications

8.8K

Citations

196

Authors

80

Institutions

Publications per year

2017–2026

18

Authors

196

  1. BR
    Bryan Russell

    Adobe Systems (United States), United States

    3 publications · 558 citations

    3

    h-index

  2. MV
    Marlène Villanova-Oliver

    Laboratoire d'Informatique de Grenoble, France

    3 publications · 98 citations

    3

    h-index

  3. SM
    Steven M. Seitz

    University of Washington, United States

    3 publications · 3.6K citations

    3

    h-index

  4. IM
    Ioan Marius Bilasco

    Université de Lille, France

    3 publications · 98 citations

    3

    h-index

  5. JG
    Jérôme Gensel

    Laboratoire d'Informatique de Grenoble, France

    3 publications · 98 citations

    3

    h-index

1–5 of 196

Venues

  1. arXiv (Cornell University)

    Repository · 4 publications · 986 citations

    4

    h-index

  2. Bioinformatics

    3 publications · 502 citations

    3

    h-index

  3. ACM Transactions on Graphics

    TOG · 3 publications · 4.2K citations

    3

    h-index

  4. 2009 IEEE Conference on Computer Vision and Pattern Recognition

    Conference · 2 publications · 98 citations

    2

    h-index

  5. Medical Image Analysis

    1 publications · 27 citations

    1

    h-index

1–5