3D-LaneNet: End-to-End 3D Multiple Lane Detection
Preprint
Abstract
We introduce a network that directly predicts the 3D layout of lanes in a road scene from a single image. This work marks a first attempt to address this task with on-board sensing without assuming a known constant lane width or relying on pre-mapped environments. Our network architecture, 3D-LaneNet, applies two new concepts: intra-network inverse-perspective mapping (IPM) and anchor-based lane representation. The intra-network IPM projection facilitates a dual-representation information flow in both regular image-view and top-view. An anchor-per-column output representation enables our end-to-end approach which replaces common heuristics such as clustering and outlier rejection, casting lane estimation as an object detection problem. In addition, our approach explicitly handles complex situations such as lane merges and splits. Results are shown on two new 3D lane datasets, a synthetic and a real one. For comparison with existing methods, we test our approach on the image-only tuSimple lane detection benchmark, achieving performance competitive with state-of-the-art.Keywords
All Related Versions
This publication has 18 references indexed in Scilit:
- EL-GAN: Embedding Loss Driven Generative Adversarial Networks for Lane Detection2019
- Deep Semantic Lane Segmentation for Mapless Driving2018
- A 3D Estimation of Structural Road Surface Based on Lane-Line InformationIFAC-PapersOnLine, 2018
- VPGNet: Vanishing Point Guided Network for Lane and Road Marking Detection and Recognition2017
- End-To-End Ego Lane Estimation Based on Sequential Transfer Learning for Self-Driving Cars2017
- You Only Look Once: Unified, Real-Time Object Detection2016
- Deep convolutional neural fields for depth estimation from a single image2015
- Robust lane detection and tracking with ransac and Kalman filter2009
- Autonomous driving in urban environments: Boss and the Urban ChallengeJournal of Field Robotics, 2008
- Multiple View Geometry in Computer Vision2004