Panoptic Segmentation

Panoptic Segmentation is a unifying image understanding task that jointly performs semantic segmentation and instance segmentation by assigning every pixel in an image to either a unique instance of a thing class or a stuff class, thereby providing a complete, consistent, and pixel-level labeling of all entities in a scene.

195

Publications

18.8K

Citations

649

Authors

160

Institutions

Publications per year

2017–2026

194

Authors

649

  1. LC
    Liang-Chieh Chen

    Google (United States), United States

    14 publications · 2.8K citations

    12

    h-index

  2. HA
    Hartwig Adam

    Google (United States), United States

    10 publications · 1.5K citations

    10

    h-index

  3. BC
    Bowen Cheng

    University of Illinois Urbana-Champaign, United States

    9 publications · 1.4K citations

    9

    h-index

  4. AK
    Alexander Kirillov

    Meta (Israel), Israel

    8 publications · 3.1K citations

    8

    h-index

  5. MD
    Maxwell D. Collins

    Google (United States), United States

    7 publications · 940 citations

    7

    h-index

1–5 of 649

Venues

  1. arXiv (Cornell University)

    Repository · 51 publications · 2.5K citations

    23

    h-index

  2. 12

    h-index

  3. IEEE Robotics and Automation Letters

    8 publications · 423 citations

    8

    h-index

  4. 6

    h-index

  5. 2021 IEEE/CVF International Conference on Computer Vision (ICCV)

    Conference · 5 publications · 517 citations

    5

    h-index

1–5