SOTAVerified

Lane Detection

Lane Detection is a computer vision task that involves identifying the boundaries of driving lanes in a video or image of a road scene. The goal is to accurately locate and track the lane markings in real-time, even in challenging conditions such as poor lighting, glare, or complex road layouts.

Lane detection is an important component of advanced driver assistance systems (ADAS) and autonomous vehicles, as it provides information about the road layout and the position of the vehicle within the lane, which is crucial for navigation and safety. The algorithms typically use a combination of computer vision techniques, such as edge detection, color filtering, and Hough transforms, to identify and track the lane markings in a road scene.

( Image credit: End-to-end Lane Detection )

Papers

Showing 1–50 of 251 papers

TitleStatusHype
Cosmos-Drive-Dreams: Scalable Synthetic Driving Data Generation with World Foundation ModelsCode3
Panacea+: Panoramic and Controllable Video Generation for Autonomous DrivingCode3
PETRv2: A Unified Framework for 3D Perception from Multi-Camera ImagesCode3
Sketch and Refine: Towards Fast and Accurate Lane DetectionCode2
TopoLogic: An Interpretable Pipeline for Lane Topology Reasoning on Driving ScenesCode2
A Keypoint-based Global Association Network for Lane DetectionCode2
Ultra Fast Deep Lane Detection with Hybrid Anchor Driven Ordinal ClassificationCode2
Ultra Fast Structure-aware Deep Lane DetectionCode2
ONCE-3DLanes: Building Monocular 3D Lane DetectionCode2
HybridNets: End-to-End Perception NetworkCode2
Monocular Lane Detection Based on Deep Learning: A SurveyCode2
DV-3DLane: End-to-end Multi-modal 3D Lane Detection with Dual-view RepresentationCode2
Anchor3DLane++: 3D Lane Detection via Sample-Adaptive Sparse 3D Anchor RegressionCode2
TopoPoint: Enhance Topology Reasoning via Endpoint Detection in Autonomous DrivingCode2
End to End Learning for Self-Driving CarsCode2
Rethinking Efficient Lane Detection via Curve ModelingCode2
Enhancing 3D Lane Detection and Topology Reasoning with 2D Lane PriorsCode2
You Only Look at Once for Real-time and Generic Multi-TaskCode2
LATR: 3D Lane Detection from Monocular Images with TransformerCode2
CLRNet: Cross Layer Refinement Network for Lane DetectionCode2
CLRerNet: Improving Confidence of Lane Detection with LaneIoUCode2
YOLOPv2: Better, Faster, Stronger for Panoptic Driving PerceptionCode2
TwinLiteNetPlus: A Stronger Model for Real-time Drivable Area and Lane SegmentationCode2
PersFormer: 3D Lane Detection via Perspective Transformer and the OpenLane BenchmarkCode2
OpenLane-V2: A Topology Reasoning Benchmark for Unified 3D HD MappingCode2
BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask LearningCode1
Augmenting Lane Perception and Topology Understanding with Standard Definition Navigation MapsCode1
3D-LaneNet: End-to-End 3D Multiple Lane DetectionCode1
Polar R-CNN: End-to-End Lane Detection with Fewer AnchorsCode1
OpenLKA: An Open Dataset of Lane Keeping Assist from Recent Car Models under Real-world Driving ConditionsCode1
DALNet: A Rail Detection Network Based on Dynamic Anchor LineCode1
CurveLane-NAS: Unifying Lane-Sensitive Architecture Search and Adaptive Point BlendingCode1
K-Lane: Lidar Lane Dataset and Benchmark for Urban Roads and HighwaysCode1
PolyLaneNet: Lane Estimation via Deep Polynomial RegressionCode1
CondLaneNet: a Top-to-down Lane Detection Framework Based on Conditional ConvolutionCode1
Lane Detection in Low-light Conditions Using an Efficient Data Enhancement : Light Conditions Style TransferCode1
LDNet: End-to-End Lane Marking Detection Approach Using a Dynamic Vision SensorCode1
CLRmatchNet: Enhancing Curved Lane Detection with Deep Matching ProcessCode1
ADNet: Lane Shape Prediction via Anchor DecompositionCode1
CLRKDNet: Speeding up Lane Detection with Knowledge DistillationCode1
An intelligent modular real-time vision-based system for environment perceptionCode1
Key Points Estimation and Point Instance Segmentation Approach for Lane DetectionCode1
Lane2Seq: Towards Unified Lane Detection via Sequence GenerationCode1
LaneAF: Robust Multi-Lane Detection with Affinity FieldsCode1
Learning to Predict Navigational Patterns from Partial ObservationsCode1
Lane Graph Estimation for Scene Understanding in Urban DrivingCode1
Gen-LaneNet: A Generalized and Scalable Approach for 3D Lane DetectionCode1
Generating Dynamic Kernels via Transformers for Lane DetectionCode1
CARLANE: A Lane Detection Benchmark for Unsupervised Domain Adaptation from Simulation to multiple Real-World DomainsCode1
FENet: Focusing Enhanced Network for Lane DetectionCode1
Show:102550
← PrevPage 1 of 6Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1DLNetF1 score81.23—Unverified
2CLRerNet-DLA34F1 score81.12—Unverified
3CLRerNet-Res101F1 score80.91—Unverified
4CondLSTR(ResNet-101)F1 score80.77—Unverified
5CLRerNet-Res34F1 score80.76—Unverified
6CLRKDNet (DLA-34)F1 score80.68—Unverified
7CLRNetV2 (DLA34)F1 score80.68—Unverified
8CondLSTR(ResNet-34)F1 score80.55—Unverified
9CLRNet(DLA-34)F1 score80.47—Unverified
10CLRNetV2 (ResNet101)F1 score80.43—Unverified
#ModelMetricClaimedVerifiedStatus
1SCNN_UNet_Attention_PL*Accuracy98.38—Unverified
2GANet(ResNet-34)F1 score97.71—Unverified
3GANet(ResNet-18)F1 score97.68—Unverified
4CLRNet(ResNet-101)F1 score97.62—Unverified
5GANet(ResNet-101)F1 score97.45—Unverified
6CondLaneNet(ResNet-34)F1 score97.01—Unverified
7CLRNetV2 (ResNet18)Accuracy96.99—Unverified
8PE-RESAAccuracy96.93—Unverified
9FOLOLane(ERFNet)Accuracy96.92—Unverified
10CLRNet(ResNet-34)Accuracy96.9—Unverified
#ModelMetricClaimedVerifiedStatus
1CondLSTR (ResNet-101)F1 score88.47—Unverified
2CondLSTR (ResNet-34)F1 score88.23—Unverified
3CondLSTR (ResNet-18)F1 score87.99—Unverified
4CANet-LF1 score87.87—Unverified
5CLRNetV2 (ResNet101)F1 score87.81—Unverified
6CANet-MF1 score87.19—Unverified
7CANet-SF1 score86.57—Unverified
8CLRerNet-DLA34F1 score86.47—Unverified
9CondLaneNet-L(ResNet-101)F1 score86.1—Unverified
10CLRNet-DLA34F1 score86.1—Unverified
#ModelMetricClaimedVerifiedStatus
1TwinLiteNetPlus-LargeIoU (%)34.2—Unverified
2TwinLiteNetPlus-MediumIoU (%)32.3—Unverified
3HybridNetsIoU (%)31.6—Unverified
4TwinLiteNetIoU (%)31.08—Unverified
5TriLiteNet-baseIoU (%)29.8—Unverified
6TwinLiteNetPlus-SmallIoU (%)29.3—Unverified
7A-YOLOM(s)IoU (%)28.8—Unverified
8YOLOPv2IoU (%)27.25—Unverified
9YOLOPIoU (%)26.2—Unverified
10TwinLiteNetPlus-NanoIoU (%)23.3—Unverified
#ModelMetricClaimedVerifiedStatus
1FENetV2mF171.85—Unverified
2CLRNet (DLA-34)F10.96—Unverified
3BézierLaneNet (ResNet-34)F10.96—Unverified
4LaneAFF10.96—Unverified
5CLRNet (ResNet-18)F10.96—Unverified
6BézierLaneNet (ResNet-18)F10.96—Unverified
7LaneATT (ResNet-34)F10.94—Unverified
8LaneATT (ResNet-122)F10.94—Unverified
9LaneATT (ResNet-18)F10.93—Unverified
10PolyLaneNetF10.88—Unverified
#ModelMetricClaimedVerifiedStatus
1DSLPIoU0.45—Unverified
2LaneGraphNetIoU0.42—Unverified
3STSUIoU0.39—Unverified
#ModelMetricClaimedVerifiedStatus
1CondLSTR (ResNet-101)F1 score63.4—Unverified
2CondLSTR (ResNet-34)F1 score62—Unverified
3CondLSTR (ResNet-18)F1 score60.1—Unverified
#ModelMetricClaimedVerifiedStatus
1VPGNetF10.88—Unverified
2Overfeat CNN detector + DBSCANF10.87—Unverified
#ModelMetricClaimedVerifiedStatus
1VPGNetF10.87—Unverified
2Overfeat CNN detector + DBSCANF10.86—Unverified
#ModelMetricClaimedVerifiedStatus
1LDNetAverage IOU62.79—Unverified
#ModelMetricClaimedVerifiedStatus
1LLDN-GFCF182.12—Unverified
#ModelMetricClaimedVerifiedStatus
1TopoLogicmAP33.2—Unverified
#ModelMetricClaimedVerifiedStatus
1SCNN_UNet_Attention_PL*F10.92—Unverified