Method

mixedCRF [la] [MixedCRF]


Submitted on 8 May. 2017 09:14 by
Xiaofeng Han (School of computer science and engineering, Nanjing University of Science and Technology, China)

Running time:6s
Environment:1 core @ 2.5 Ghz (Matlab + C/C++)

Method Description:
CRF based on liadr point clouds and images
Parameters:
Latex Bibtex:

Evaluation in Bird's Eye View


Benchmark MaxF AP PRE REC FPR FNR
UM_ROAD 91.57 % 84.68 % 90.02 % 93.19 % 4.71 % 6.81 %
UMM_ROAD 92.75 % 90.24 % 94.03 % 91.50 % 6.39 % 8.50 %
UU_ROAD 85.69 % 75.12 % 80.17 % 92.02 % 7.42 % 7.98 %
URBAN_ROAD 90.59 % 84.24 % 89.11 % 92.13 % 6.20 % 7.87 %
This table as LaTeX

Behavior Evaluation


This table as LaTeX

Road/Lane Detection

The following plots show precision/recall curves for the bird's eye view evaluation.


Distance-dependent Behavior Evaluation

The following plots show the F1 score/Precision/Hitrate with respect to the longitudinal distance which has been used for evaluation.


Visualization of Results

The following images illustrate the performance of the method qualitatively on a couple of test images. We first show results in the perspective image, followed by evaluation in bird's eye view. Here, red denotes false negatives, blue areas correspond to false positives and green represents true positives.



This figure as: png

This figure as: png

This figure as: png

This figure as: png

This figure as: png

This figure as: png

This figure as: png

This figure as: png

This figure as: png

This figure as: png

This figure as: png

This figure as: png

This figure as: png

This figure as: png

This figure as: png

This figure as: png

This figure as: png

This figure as: png


eXTReMe Tracker