[1] proposed to improve training of two-stage detectors by constructing minibatches using high-loss examples. Specifically, in OHEM each example is scored by its loss, non-maximum suppression (nms) is then applied, and a minibatch is constructed with the highest-loss examples. The nms threshold and batch size are tunable parameters. Like the focal loss, OHEM puts more emphasis on misclassified examples, but unlike FL, OHEM completely discards easy examples.
[1] A. Shrivastava, A. Gupta, and R. Girshick. Training regionbased object detectors with online hard example mining. In CVPR, 2016. 2, 3, 6, 7
