F1: 95.6X error near: 0.052X error far: 0.306Z error near: 0.015
Papers with Code · CC BY-SA 4.0
3D Lane Detection
The goal of 3D Lane Detection is to perceive lanes that provide guidance for autonomous vehicles. A lane can be represented as a visible laneline or a conceptual centerline. Furthermore, a lane obtains extra attributes from the understanding of the surrounding environment.
<span style="color:grey; opacity: 0.6">( Image credit: OpenLane-V2 )</span>
개요는 각 항목의 일부입니다. 위 탭에서 전체 기록을 페이지별로 탐색할 수 있습니다.
벤치마크
Papers with Code 아카이브의 데이터셋별 기록 예시입니다. 평가 조건과 지표 방향이 검증되지 않아 순위를 의미하지 않습니다.
Dataset기록된 모델 예시MetricPaperCode
데이터셋
4개논문
F1: 86.4X error near: 0.068X error far: 0.477Z error near: 0.015
F1: 96.9X error near: 0.016X error far: 0.242Z error near: 0.02
F1: 95.8X error near: 0.078X error far: 0.326Z error near: 0.018
F1: 88.1X error near: 0.061X error far: 0.496Z error near: 0.012
F1 (all): 66.0Up & Down: 61.7Curve: 72.7Extreme Weather: 63.8
F1: 97.4X error near: 0.030X error far: 0.277Z error near: 0.011
F1: 91.9X error near: 0.061X error far: 0.361Z error near: 0.029
F1: 92.9X error near: 0.054X error far: 0.356Z error near: 0.010
F1: 91.9X error near: 0.049X error far: 0.387Z error near: 0.008
OLS: 39.2DET_l: 22.1DET_t: 70.6TOP_ll: 6.0
라이브러리
Papers with Code 리더보드 row에 연결된 구현 저장소입니다.