Pedestrian lane detection in unstructured scenes for assistive navigation
作者：
Highlights：
• We propose an algorithm for pedestrian lane detection in unstructured scenes.
• Vanishing point is found via color tensor and local orientations of edge pixels.
• Sample lane region is then located via color and geometrics features of lane.
• A lane model adaptive to the input image is constructed for lane segmentation.
• We create a dataset for vanishing point estimation and pedestrian lane detection.
摘要
•We propose an algorithm for pedestrian lane detection in unstructured scenes.•Vanishing point is found via color tensor and local orientations of edge pixels.•Sample lane region is then located via color and geometrics features of lane.•A lane model adaptive to the input image is constructed for lane segmentation.•We create a dataset for vanishing point estimation and pedestrian lane detection.
论文关键词：Pedestrian lane detection,Vanishing point estimation,Assistive and autonomous navigation,Benchmark dataset
论文评审过程：Received 15 April 2015, Revised 31 December 2015, Accepted 23 January 2016, Available online 7 June 2016, Version of Record 7 June 2016.
论文官网地址：https://doi.org/10.1016/j.cviu.2016.01.011
