GIoU 损失函数的局限性：预测框在真实框内时的位置信息获取
As shown in Figure 5, although the GIoU loss function solves the problem of non-intersecting between the true bounding box and the predicted bounding box, when the predicted box is within the true box, i.e., A∩B=B, it cannot obtain the position information of the predicted box well.
原文地址: https://www.cveoy.top/t/topic/nimQ 著作权归作者所有。请勿转载和采集!