Weakly Supervised Semantic Segmentation

Weakly Supervised Semantic Segmentation is a semantic segmentation approach that belongs to the category of supervised learning methods but is differentiated by its use of incomplete, limited, or noisy supervision—such as image-level labels, bounding boxes, or scribbles—instead of dense pixel-level annotations for training.

371

Publications

32.4K

Citations

1.3K

Authors

358

Institutions

Publications per year

2017–2026

335

Authors

1.3K

  1. YW
    Yunchao Wei

    Beijing Jiaotong University, China

    16 publications · 4K citations

    14

    h-index

  2. MC
    Ming‐Ming Cheng

    Nankai University, China

    10 publications · 3K citations

    10

    h-index

  3. GL
    Guosheng Lin

    Nanyang Technological University, Singapore

    10 publications · 569 citations

    10

    h-index

  4. ZZ
    Zhaoxiang Zhang

    Chinese Academy of Sciences, China

    9 publications · 745 citations

    9

    h-index

  5. JX
    Jimin Xiao

    Xi’an Jiaotong-Liverpool University, China

    8 publications · 552 citations

    8

    h-index

1–5 of 1.3K

Venues

  1. arXiv (Cornell University)

    Repository · 68 publications · 2.4K citations

    27

    h-index

  2. 18

    h-index

  3. IEEE Transactions on Pattern Analysis and Machine Intelligence

    TPAMI · 18 publications · 1.8K citations

    17

    h-index

  4. Proceedings of the AAAI Conference on Artificial Intelligence

    Conference · 16 publications · 1.4K citations

    15

    h-index

  5. IEEE Transactions on Image Processing

    13 publications · 1.7K citations

    12

    h-index

1–5