SwiftLane: Towards Fast and Efficient Lane Detection

@article{Jayasinghe2021SwiftLaneTF,
  title={SwiftLane: Towards Fast and Efficient Lane Detection},
  author={Oshada Jayasinghe and Damith Anhettigama and Sahan Hemachandra and Shenali Kariyawasam and Ranga Rodrigo and Peshala G. Jayasekara},
  journal={2021 20th IEEE International Conference on Machine Learning and Applications (ICMLA)},
  year={2021},
  pages={859-864}
}
Recent work done on lane detection has been able to detect lanes accurately in complex scenarios, yet many fail to deliver real-time performance specifically with limited computational resources. In this work, we propose SwiftLane: a simple and light-weight, end-to-end deep learning based framework, coupled with the row-wise classification formulation for fast and efficient lane detection. This framework is supplemented with a false positive suppression algorithm and a curve fitting technique… 

Figures and Tables from this paper

PersFormer: 3D Lane Detection via Perspective Transformer and the OpenLane Benchmark

It is shown that PersFormer significantly outperforms competitive baselines in the 3D lane detection task on the authors' new OpenLane dataset as well as Apollo 3D Lane Synthetic dataset, and is also on par with state-of-the-art algorithms in the 2D task on OpenL Lane.

Perspective Transformer and MobileNets-Based 3D Lane Detection from Single 2D Image

This work has put forward a high-quality method to predict 3D lanes from a single 2D image captured by conventional cameras, which is also cost effective and requires no exceptional computational resources, thereby reducing its implementation costs.

Sharpening and Detection of Road Studs for Intelligent Vehicles at Nighttime

This research designs a robust approach for detecting road studs at nighttime, which is the combination of various statistical methods and presents that the designed approach is accurate and robust for road stud detection against nighttime.

Opendenselane: A Dense Lidar-Based Dataset for HD Map Construction

In autonomous driving system, High-Definition (HD) map is an important basis for localization, perception and planning tasks. For the construction of HD map, land marking detection is the first step.

References

SHOWING 1-10 OF 26 REFERENCES

Ultra Fast Structure-aware Deep Lane Detection

A novel, simple, yet effective formulation aiming at extremely fast speed and challenging scenarios, which treats the process of lane detection as a row-based selecting problem using global features and proposes a structural loss to explicitly model the structure of lanes.

Keep your Eyes on the Lane: Real-time Attention-guided Lane Detection

This work proposes an anchor-based deep lane detection model, which, akin to other generic deep object detectors, uses the anchors for the feature pooling step, and shows that the method outperforms the current state-of-the-art methods showing both higher efficacy and efficiency.

End-to-End Lane Marker Detection via Row-wise Classification

This paper proposes a method performing direct lane marker vertex prediction in an end-to-end manner, i.e., without any post-processing step that is required in the pixel-level dense prediction task, and translates the lane marker detection problem into a row-wise classification task, which takes advantage of the innate shape of lane markers.

RESA: Recurrent Feature-Shift Aggregator for Lane Detection

A novel module named REcurrent Feature-Shift Aggregator (RESA) to enrich lane feature after preliminary feature extraction with an ordinary CNN and achieves state-of-the-art results on two popular lane detection benchmarks (CULane and Tusimple).

PathMark: A Novel Fast Lane Detection Algorithm for Embedded Systems

A new, fast and robust lane detection algorithm based on searching lane marker candidates by the brightness feature and geometric matching from discrete segments to full lane markers, and then followed by a random noise filter that is more suitable for real safety driving use in embedded systems.

Key Points Estimation and Point Instance Segmentation Approach for Lane Detection

A traffic line detection method called Point Instance Network (PINet), based on the key points estimation and instance segmentation approach, which achieves competitive accuracy and false positive on CULane and TuSimple datasets, popular public datasets for lane detection.

Focus on Local: Detecting Lane Marker from Bottom Up via Key Point

This work proposes a novel lane marker detection solution, FOLOLane, that greatly outperforms all existing methods on public datasets while achieving the best state-of-the-art results and real-time processing simultaneously.

CurveLane-NAS: Unifying Lane-Sensitive Architecture Search and Adaptive Point Blending

A novel lane-sensitive architecture search framework named CurveLane-NAS to automatically capture both long-ranged coherent and accurate short-range curve information while unifying both architecture search and post-processing on curve lane predictions via point blending.

Road Lane Detection Using H-Maxima and Improved Hough Transform

  • K. GhazaliRui XiaoJie Ma
  • Computer Science
    2012 Fourth International Conference on Computational Intelligence, Modelling and Simulation
  • 2012
A fast and improved algorithm with the ability to detect unexpected lane changes is aimed in this paper which first defines the region of interest from input image for reducing searching space, and divided the image into near field of view and far field of views.

Real-time lane detection and departure warning system on embedded platform

  • Youngwan LeeHakil Kim
  • Computer Science
    2016 IEEE 6th International Conference on Consumer Electronics - Berlin (ICCE-Berlin)
  • 2016
Results show good performance with an average correct detection rate of 96% under various challenging urban and highway conditions while the processing time takes only 22.76 ms per frame on the embedded board which verifies that the proposed method could be feasible for real-time applications in commercial ADAS products.