TL;DR本篇论文探讨了基于 YOLOX 算法的 DecIoU 边界框回归损失函数和 Push Loss 用于改进路面物体检测问题,并且通过采用动态锚箱机制提升了检测精度,实验在 KITTI 数据集上证明了该方法的有效性。
Abstract
road object detection is an important branch of automatic driving technology,
The model with higher detection accuracy is more conducive to the safe driving
of vehicles. In road object detection, the omission of