« View All Resources

Multi-Target Tracking Annotations and Detections

- Posted on December 15, 2011 at 7:24 am.

  • Links: http://www.gris.informatik.tu-darmstadt.de/~aandriye/data.html
  • Details:
  • Data

    Download all data: data-tud.zip (2.3 MB)

    Ground Truth

    All annotations were done by manually placing bounding boxes around pedestrians and interpolating their trajectories between key frames. The trajectories were then smoothed to avoid discontinuous or jittery movements. All targets were annotated in all sequences, even in case of total occlusion. Each person entering the field of view acquires a unique ID, i.e. if a person leaves the screen and later reappears again, a new ID is assigned. Please note that bounding boxes are not always perfectly aligned due to articulation, interpolation and mistakes made by the annotator. 
    The annotations are saved in xml files using theCVML Specification, similar to the ground truth of the CAVIAR dataset
    For each dataset there are two ground truth files. One (complete) contains annotations for all visible targets, the other one (cropped) only contains targets within the predefined tracking area used in our experiments. 
    Please cite our work if you use the provided ground truth.

    PETS 2009

    All annotations were done in View001. The PETS 2009 dataset is available here
    PETS-S2L1 Ground Truth 
    Tracking Area in world coordinates (xmin, xmax, ymin, ymax): (-140696, 49813, -142740, 17335) 

    S2.L1 (12-34)

    S2.L2 (14-55)

    S2.L3 (14-41)

    S1.L1-1 (13-57)

    S1.L1-2 (13-59)

    S1.L2-1 (14-06)

    S1.L2-2 (14-31)

    complete
    cropped
    complete
    cropped
    complete
    cropped
    complete
    cropped
    complete
    cropped
    complete
    cropped
    complete
    cropped


    TUD-Stadtmitte

    We here provide our own annotations for the widely used TUD-Stadtmitte sequence, where bounding boxes for all targets inside the field of view (even the ones entirely occluded) are annotated. The TUD dataset is available here

    TUD-Stadtmitte Ground Truth 
    Tracking Area in world coordinates (xmin, xmax, ymin, ymax): (-19, 12939, -48, 10053) 

    TUD-Stadtmitte


    The camera calibration for the TUD-Stadtmitte sequence is here. It has the same format as the calibration for the PETS dataset


    TUD-Campus and TUD-Crossing

    We extend the original ground truth for these two sequences by providing correct association of people identities across frames. The bounding boxes positions and sizes are kept unchanged. Additionally, we linearly interpolate the annotations through occlusions, such that, again, all targets are annotated, regardless of their visibility fraction.
    TUD-Campus TUD-Crossing

    TUD-Campus

    TUD-Crossing

    original 
    interpolated
    original 
    interpolated


    k();} ?>