Method

ZongNet [ZongNet]
[Anonymous Submission]

Submitted on 11 Jul. 2018 08:10 by
[Anonymous Submission]

Running time:0.1 s
Environment:1 core @ 2.5 Ghz (C/C++)

Method Description:
SeNet+Modified DenseFCN + Feature pyramid fusion + Data enhancement
Parameters:
308M
Latex Bibtex:

Evaluation in Bird's Eye View


Benchmark MaxF AP PRE REC FPR FNR
UM_ROAD 96.70 % 90.12 % 96.00 % 97.41 % 1.85 % 2.59 %
UMM_ROAD 97.53 % 93.05 % 97.12 % 97.94 % 3.19 % 2.06 %
UU_ROAD 95.08 % 85.88 % 92.01 % 98.36 % 2.78 % 1.64 %
URBAN_ROAD 96.68 % 90.05 % 95.50 % 97.88 % 2.54 % 2.12 %
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