Method

Combination of 2D and 3D Features [st] [Feature++]


Submitted on 30 Jul. 2016 16:18 by
wenli he (xiamen university)

Running time:13 s
Environment:4 core @ 2.5 Ghz (Matlab + C/C++)

Method Description:
Combination of 2D and 3D Features. Features consist of
RGB、Gabor、GKDES、Pos3D and Plane.
Parameters:
superpixels=1000
Latex Bibtex:

Evaluation in Bird's Eye View


Benchmark MaxF AP PRE REC FPR FNR
UM_ROAD 90.43 % 88.83 % 89.63 % 91.26 % 4.81 % 8.74 %
UMM_ROAD 93.55 % 92.34 % 92.77 % 94.34 % 8.08 % 5.66 %
UU_ROAD 87.40 % 85.48 % 86.19 % 88.64 % 4.63 % 11.36 %
URBAN_ROAD 91.12 % 89.51 % 90.16 % 92.10 % 5.54 % 7.90 %
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.



This figure as: png eps pdf

This figure as: png eps pdf

This figure as: png eps pdf

This figure as: png eps pdf

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