目录
SUPER: A Novel Lane Detection System
LaneAF: Robust Multi-Lane Detection with Affinity Fields
Structure Guided Lane Detection
持续更新中......
2020年
SUPER: A Novel Lane Detection System
对语义分割模型采用层级分类,在推理过程中,每个子分类器根据自己的决策由其父分类器控制。通过这种方式,获得了更可靠的场景标签。在Apollo数据集中,车道标记标签比地面和目标标签更准确,因此在训练阶段只使用车道分割标签。
然而,车道标记附近的一些地面区域很容易被误认为车道标记。为了解决这一问题,我们将车道标记扩展到附近,并将整个扩展区域标记为“道路”。被车辆等物体阻塞的道路区域将从标签中移除。下图展示了Apollo标签在扩展处理之前(左)和之后(右)的示例。车道标记和生成的“道路”标签都用于训练。
可参考:https://mp.weixin.qq.com/s/t-Nf6cttegYkBLEKGGJfSQ
2021年
LaneAF: Robust Multi-Lane Detection with Affinity Fields
类似于laneNet,学习车道线的embed,只是这个embed做了特殊处理,直接用车道线gt的几何特征来监督;
Structure Guided Lane Detection
CondLaneNet
其他车道线检测论文集合
1、Efficient Road Lane Marking Detection with Deep Learning
2、VPGNet: Vanishing Point Guided Network for Lane and Road Marking Detection and Recognition
3、Semi-Local 3D Lane Detection and Uncertainty Estimation
4、Robust Lane Marking Detection Algorithm Using Drivable Area Segmentation and Extended SLT
5、3D-LaneNet: End-to-End 3D Multiple Lane Detection
7、Robust Lane Detection from Continuous Driving Scenes Using Deep Neural Networks
8、End to End Video Segmentation for Driving : Lane Detection For Autonomous Car
9、Multiple Lane Detection Algorithm Based on Optimised Dense Disparity Map Estimation
10、Real-time stereo vision-based lane detection system
11、LaneNet: Real-Time Lane Detection Networks for Autonomous Driving
12、Towards End-to-End Lane Detection: an Instance Segmentation Approach
13、A Robust Lane Detection and Departure Warning System
14、Real time Detection of Lane Markers in Urban Streets
15、LaNet: Real-time Lane Identification by Learning Road SurfaceCharacteristics from Accelerometer Data
17、Key Points Estimation and Point Instance Segmentation Approach for Lane Detection
19、End-to-end Lane Detection through Differentiable Least-Squares Fitting
20、Real-time Lane Marker Detection Using Template Matching with RGB-D Camera
21、LIDAR-Camera Fusion for Road Detection Using Fully Convolutional Neural Networks