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

facebooktwittermailshare

Parking Space Detection Based on A Multi-task Deep Convolutional Network with Spatial Transform

Abstract: 

Vacant parking space detection is a challenging vision task due to outdoor lighting variation and perspective distortion. Previous methods found on camera geometry and projection matrix to select space image region for status classification. By utilizing suitable hand-crafted features, outdoor lighting variation and perspective distortion could be well handled. However, if also considering parking displacement, non-unified car size, and inter-object occlusion, we find the problem becomes more troublesome. To overcome these problems, we propose a deep learning framework to infer the parking status with two contributions. First, we integrate a convolutional spatial transformer network (STN) to crop the local image area adaptively according to car size and parking displacement. Second, in order to solve inter-object occlusion problems, we group 3 neighboring spaces as a unit. A multi-task loss function is designed to consider the status estimation of the target space and its two neighbors jointly. With the loss function, we could force our network to learn occlusion patterns while estimating space status. The results show our system can reduce the error detection rate and thereby increase system accuracy.

up
0 users have voted:

Paper Details

Authors:
Ching-Chun Huang
Submitted On:
8 September 2017 - 5:24am
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Hoang Tran Vu
Paper Code:
1711
Document Year:
2017
Cite

Document Files

Parking-Space-Detection-Based-on-A-Multi-task-Deep-Convolutional-Network-with-Spatial-Transform.pdf

(7 downloads)

Subscribe

[1] Ching-Chun Huang, "Parking Space Detection Based on A Multi-task Deep Convolutional Network with Spatial Transform", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1874. Accessed: Sep. 21, 2017.
@article{1874-17,
url = {http://sigport.org/1874},
author = {Ching-Chun Huang },
publisher = {IEEE SigPort},
title = {Parking Space Detection Based on A Multi-task Deep Convolutional Network with Spatial Transform},
year = {2017} }
TY - EJOUR
T1 - Parking Space Detection Based on A Multi-task Deep Convolutional Network with Spatial Transform
AU - Ching-Chun Huang
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1874
ER -
Ching-Chun Huang. (2017). Parking Space Detection Based on A Multi-task Deep Convolutional Network with Spatial Transform. IEEE SigPort. http://sigport.org/1874
Ching-Chun Huang, 2017. Parking Space Detection Based on A Multi-task Deep Convolutional Network with Spatial Transform. Available at: http://sigport.org/1874.
Ching-Chun Huang. (2017). "Parking Space Detection Based on A Multi-task Deep Convolutional Network with Spatial Transform." Web.
1. Ching-Chun Huang. Parking Space Detection Based on A Multi-task Deep Convolutional Network with Spatial Transform [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1874