log average miss rate

Log average miss rate (LAMR) is a metric used to evaluate the performance of object detection models. It is calculated by taking the logarithm of the average miss rate (AMR) over different false positive rates (FPR) between 0 and 1. The AMR is the average number of missed detections per image, and the FPR is the fraction of false positive detections per image.

LAMR is a useful metric because it provides a single number that summarizes the overall performance of a detection model across all possible FPR values. It is commonly used in benchmark datasets such as PASCAL VOC and COCO to compare the performance of different object detection models.

A lower LAMR indicates better performance, as it means the model has fewer missed detections and false positives across a range of FPR values.