This is a fork of official code for "Structured Bird’s-Eye-View Traffic Scene Understanding from Onboard Images"
Original Code - https://github.com/ybarancan/STSU
In this work, we enhance the multi view lane line extraction represented in BEV coordinates using a multi view camera images instead of single camera image as in original work
All the instructions are exactly same as the one mentioned in the original paper. We enhanced the ground truth generation and additional encoder components to get multi view representation