Joint Feature Embedding

Joint Feature Embedding is a multimodal fusion method that learns a shared latent representation space by simultaneously integrating features from multiple modalities through coordinated transformations, thereby preserving cross-modal correlations while enabling unified downstream processing.

539

Publications

37.6K

Citations

2K

Authors

591

Institutions

Publications per year

2017–2026

464

Authors

2K

  1. BS
    Bernt Schiele

    Max Planck Institute for Informatics, Germany

    11 publications · 2.6K citations

    11

    h-index

  2. MW
    Meng Wang

    Hefei University of Technology, China

    9 publications · 501 citations

    9

    h-index

  3. ZA
    Zeynep Akata

    Max Planck Institute for Informatics, Germany

    8 publications · 1.9K citations

    8

    h-index

  4. DS
    Dinggang Shen

    University of North Carolina at Chapel Hill, United States

    5 publications · 1.7K citations

    5

    h-index

  5. DG
    Dan Guo

    Hefei University of Technology, China

    5 publications · 317 citations

    5

    h-index

1–5 of 2K

Venues

  1. arXiv (Cornell University)

    Repository · 55 publications · 2.2K citations

    26

    h-index

  2. Pattern Recognition

    31 publications · 1.4K citations

    23

    h-index

  3. Neurocomputing

    22 publications · 550 citations

    16

    h-index

  4. IEEE Transactions on Image Processing

    15 publications · 995 citations

    14

    h-index

  5. Knowledge-Based Systems

    16 publications · 1K citations

    13

    h-index

1–5