In case of using a pretrained YOLOv3 object detector, the anchor boxes calculated on that particular training dataset need to be specified.
確定! 回上一頁