For this competition track, we ask the participants to perform human detection in the thermal modality. Thermal cameras provide temperature readings from the scene. They are less noisy than depth cameras, but at a comparable price they offer a much lower image resolution.
Given the provided thermal images (and bounding box groundtruth annotations), the participants will be asked to develop their thermal-based human detection method. The method will need to output a list of bounding boxes (along with associated confidence scores) per frame containing each person in it. The performance the image-based human detection methods will be evaluated in terms of average precision.