Settings

Now showing1 - 2 of 2
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Research Data, Access status: with files ,
    3D Zebrafish Tracking Dataset
    (2026-07-02)
    Neidhardt, Maximilian
    ;
    Bhattacharya, Debayan
    This dataset contains synchronized stereo-camera RGB recordings of zebrafish swimming with varying group sizes (2, 5, 10 fish), along with manually annotated frames for tracking and evaluation. It supports training and benchmarking of deep learning–based multi-object tracking and classification. The dataset includes raw video data, annotations, model weights, and example predictions. The data processing is described in detail in: Neidhardt, Maximilian, et al. "Robust Motion Tracking and Classification of Zebrafish with Deep Learning." IEEE Access (2026).
    Data Type: Video
      112  66
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Research Data, Access status: with files ,
    Position Data of Offline Needle Steering
    (2022-10-07)
    Neidhardt, Maximilian
    ;
    Reinecke, Anton
    ;
    Rogalla, Antje
    ;
    Lehmann, Sascha
    ;
    Sprenger, Johanna
    "Automated needle steering in soft tissue" is an open problem in both surgery robotics and formal verification. Challenges that are common to both fields include the interaction between needle and tissue, which depends on the coupled effects of needle deflection, friction and force, and tissue deformation; navigation, collision avoidance, and path planning, which are faced with inhomogeneous, partially unknown tissue; and also adaptive updates and feedback, which are required in real time. For each experiment, we record the feed of the needle in millimeters, the x, y, and z position of the needle in millimeters, and the rotation angle of the alignment of the angled needle tip in degree. The setup allows the needle to be pushed forwards and backward in the y direction. The experiments we restrict to a two-dimensional setting, but three-dimensional position data are recorded. Accordingly, we execute only motion plans with half rotations by 180 degree rotations. We do not execute motion plans with backward movement. Also, we do not allow more than two rotations within one motion plan.
    Data Type: Dataset
      167  154