1 paper
Junhong Xiang, Jingmin Zhang, Zhixiong Nan
To handle the two shortcomings of existing methods, (i)nearly all models rely on high-definition (HD) maps, yet the map information is not always available in real traffic scenes a…