Semi-direct Visual Slam

Semi-Direct Visual SLAM is a visual simultaneous localization and mapping method that estimates camera motion and scene structure by partially leveraging pixel intensity information without explicit feature extraction, distinguishing itself from direct methods through selective use of photometric error minimization and from indirect methods by avoiding full feature-based matching.

21

Publications

871

Citations

76

Authors

29

Institutions

Publications per year

2017–2026

17

Authors

76

  1. XG
    Xiang Gao

    Tsinghua University, China

    2 publications · 156 citations

    2

    h-index

  2. RW
    Rui Wang

    Technical University of Munich, Germany

    2 publications · 119 citations

    2

    h-index

  3. NY
    Nan Yang

    Technical University of Munich, Germany

    2 publications · 119 citations

    2

    h-index

  4. JC
    Javier Civera

    Universidad de Zaragoza, Spain

    2 publications · 105 citations

    2

    h-index

  5. DC
    Daniel Cremers

    Technical University of Munich, Germany

    2 publications · 119 citations

    2

    h-index

1–5 of 76

Venues

  1. IEEE Robotics and Automation Letters

    3 publications · 182 citations

    3

    h-index

  2. Robotics and Autonomous Systems

    2 publications · 66 citations

    2

    h-index

  3. Journal of Intelligent & Robotic Systems

    1 publications · 15 citations

    1

    h-index

  4. 1

    h-index

  5. Computers & Electrical Engineering

    1 publications · 11 citations

    1

    h-index

1–5