Evaluation Metrics :-

Qualitative Precision (QP): QP is a performance metric used to assess the effectiveness of an object tracking model at night by combining the quality of a video frame and the center location error (CLE) in object tracking at that particular frame. The Image Quality Assessment (IQA) metric assesses the visibility of individual frames in a video sequence which ranges from 0 to 1, whereas center location error is the Euclidean distance between the center of the ground truth and the predicted bounding box. The frame will be considered a positive frame when the multiplication of IQA and the CLE is less than a threshold (15.0-pixel distance). Mathematically, the QP can be expressed as:

The number of positive frames is the count of frames where the multiplication between IQA and the CLE of the frame is below the specified threshold (15.0), which is expressed as:

The total number of frames in the video is the total number of frames in the video sequence.

The Image Quality Assessment (IQA) for this competition is calculated as,

Where the Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE) score ranges from 0 to 100. Lower BRISQUE scores correspond to better perceptual image quality (clearer, less distorted frames), while higher scores indicate poorer quality (more degraded or noisy frames). Specifically, scores of 0–20 represent excellent quality, 20–40 good, 40–60 fair, 60–80 poor, and 80–100 very poor quality. By normalizing and inverting the BRISQUE score in this way, IQA is mapped to a 0–1 range where higher values correspond to higher visibility, making it compatible with the QP metric, which combines frame visibility and tracking accuracy measured via the CLE.

The value of QP will vary between zero and unity. The unit value of QP indicates that the tracker algorithm is completely successful in tracking the object, whereas the zero value of QP indicates that the algorithm failed to track the object. This metric provides a more comprehensive evaluation of tracking performance than Precision alone at night, as it considers how the visibility in the frame degrades at night.