Robust Lane Detection and Tracking in Challenging Scenarios

By Z. Kim. IEEE transactions on intelligent transportation systems (Print), 2008

592 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Automotive Engineering, Signal Processing

View Paper: https://www.semanticscholar.org/paper/84e3e701acc1e9a21086fb182ab83b8905f3585f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Automotive Engineering

In Signal Processing

Citations in the corpus

Deep Neural Network for Structural Prediction and Lane Detection in Traffic Scene (2017, methodology): “Optionally, higher level models (such as those in [6]–[8]) of the lane structures can be constructed based on the predictions.”
A Sensor-Fusion Drivable-Region and Lane-Detection System for Autonomous Vehicle Navigation in Challenging Road Scenarios (2014, methodology): “For more challenging structured roads, Kim [16] introduced a robust real-time lane detection algorithm based on Random Sample Consensus (RANSAC), and this was combined with a particle-ﬁltering-based tracking algo-rithm using a probabilistic grouping framework.”
Overview of Environment Perception for Intelligent Vehicles (2017, methodology): “In all spline models [12], the curves were parameterized using a set of control points either on [47] or near [48] the curve.”
Computing Systems for Autonomous Driving: State of the Art and Challenges (2020, methodology): “This information is usually combined with a Hough transform [91], [92] and particle or Kalman filters [89], [93], [94] to detect lane markings.”
Probabilistic Lane Tracking in Difficult Road Scenarios Using Stereovision (2009, unknown): “In [12], the authors described a system that uses a hybrid approach, i.”
CondLaneNet: a Top-to-down Lane Detection Framework Based on Conditional Convolution (2021, methodology): “Traditional lane detection methods usually rely on handcrafted operators to extract features [24, 43, 13, 17, 15, 1, 16, 33], and then fit the line shape through post-processing such as Hough transform [24, 43] and Random Sampling Consensus (RANSAC) [17, 15].”
Towards End-to-End Lane Detection: an Instance Segmentation Approach (2018, background): “, which are possibly combined with a hough transform [23], [37] and particle or Kalman filters [18], [8], [34].”
Perception, Planning, Control, and Coordination for Autonomous Vehicles (2017, methodology): “Zhang et al. [64] proposed an SVM (Support Vector Machine) based classifier to classify the clusters into ground, vegetation, building, power line, and vehicle.”
Line-CNN: End-to-End Traffic Line Detection With Line Proposal Unit (2020, methodology): “In the second “segment clustering” step, the detected line fragments are allocated into traffic lines by certain clustering methods such as DBSCAN [7], [45], RANSAC [6], [46], [47] line fitting or some heuristic greedy selections as in [10], [11], and [43].”
Variational Autoencoder for End-to-End Control of Autonomous Driving with Novelty Detection and Training De-biasing (2018, background): “These submodules can range from mapping and localization [3], [4], to perception [5]–[7], planning [8], [9], and control [10], [11].”
Data-Driven Intelligent Transportation Systems: A Survey (2011, methodology): “patches, followed by utilizing particle-filtering-based tracking algorithm to track the lanes [27].”
Finding multiple lanes in urban road networks with vision and lidar (2009, unknown): “We also eliminate substantial computational expense by operating directly on raw camera images rather than on inverse-perspective corrected images (McCall and Trivedi 2006; Kim 2008), while still reaping the benefits of calibrated cameras and realworld measurements.”
Recent progress in road and lane detection: a survey (2014, methodology): “Most methods use either The Kalman Filter [4] [3] [10] [53] or Particle Filtering [8] [22] [53] [9] for this task.”