Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Deep CNN with colorLines model for unmarked road segmentation

Abstract: 

Road detection from a monocular camera is a perception module in any advanced driver assistance or autonomous driving system. Traditional techniques work reasonably well for this problem when the roads are well maintained and the boundaries are clearly marked. However, in many developing countries or even for the rural areas in the developed countries, the assumption does not hold which leads to failure of such techniques. In this paper, we propose a novel technique based on the combination of deep convolutional neural networks (CNNs), along with color lines model based prior in a conditional random field (CRF) framework. While the CNN learns the road texture, the color lines model allows adapting to varying illumination conditions. We show that our technique outperforms the state of the art segmentation techniques on the unmarked road segmentation problem. Though, not a focus of this paper, we show that even on the standard benchmark datasets like KITTI and CamVid, where the road boundaries are well marked, the proposed technique performs competitively to the contemporary techniques.

up
0 users have voted:

Paper Details

Authors:
Shashank Yadav, Suvam Patra, Chetan Arora, Subhashis Banerjee
Submitted On:
15 September 2017 - 4:09am
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Shashank Yadav
Paper Code:
MQ-L3.3
Document Year:
2017
Cite

Document Files

Deep CNN with colorLines model for unmarked road segmentation.pdf

(92 downloads)

Subscribe

[1] Shashank Yadav, Suvam Patra, Chetan Arora, Subhashis Banerjee, "Deep CNN with colorLines model for unmarked road segmentation", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2099. Accessed: Jun. 18, 2018.
@article{2099-17,
url = {http://sigport.org/2099},
author = {Shashank Yadav; Suvam Patra; Chetan Arora; Subhashis Banerjee },
publisher = {IEEE SigPort},
title = {Deep CNN with colorLines model for unmarked road segmentation},
year = {2017} }
TY - EJOUR
T1 - Deep CNN with colorLines model for unmarked road segmentation
AU - Shashank Yadav; Suvam Patra; Chetan Arora; Subhashis Banerjee
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2099
ER -
Shashank Yadav, Suvam Patra, Chetan Arora, Subhashis Banerjee. (2017). Deep CNN with colorLines model for unmarked road segmentation. IEEE SigPort. http://sigport.org/2099
Shashank Yadav, Suvam Patra, Chetan Arora, Subhashis Banerjee, 2017. Deep CNN with colorLines model for unmarked road segmentation. Available at: http://sigport.org/2099.
Shashank Yadav, Suvam Patra, Chetan Arora, Subhashis Banerjee. (2017). "Deep CNN with colorLines model for unmarked road segmentation." Web.
1. Shashank Yadav, Suvam Patra, Chetan Arora, Subhashis Banerjee. Deep CNN with colorLines model for unmarked road segmentation [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2099