Journal article

SOCHOR Jakub, ŠPAŇHEL Jakub and HEROUT Adam. BoxCars: Improving Fine-Grained Recognition of Vehicles using 3-D Bounding Boxes in Traffic Surveillance. IEEE Transactions on Intelligent Transportation Systems. 2018, vol. 2019, no. 1, pp. 97-108. ISSN 1524-9050. Available from: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8307405
Publication language:english
Original title:BoxCars: Improving Fine-Grained Recognition of Vehicles using 3-D Bounding Boxes in Traffic Surveillance
Title (cs):BoxCars: Vylepšení rozpoznání typů vozidel s využitím 3D boxů v dopravním dohledu
Pages:97-108
Place:US
Year:2018
URL:https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8307405
Journal:IEEE Transactions on Intelligent Transportation Systems, Vol. 2019, No. 1, US
ISSN:1524-9050
DOI:10.1109/TITS.2018.2799228
Keywords
fine-grained recognition, traffic surveillance, 3D bounding boxes, convolutional neural networks
Annotation
In this paper, we focus on fine-grained recognition of vehicles mainly in traffic surveillance applications. We propose an approach that is orthogonal to recent advancements in fine-grained recognition (automatic part discovery, bilinear pooling). Also, in contrast to other methods focused on fine-grained recognition of vehicles, we do not limit ourselves to a frontal/rear viewpoint, but allow the vehicles to be seen from any viewpoint. Our approach is based on 3D bounding boxes built around the vehicles. The bounding box can be automatically constructed from traffic surveillance data. For scenarios where it is not possible to use precise construction, we propose a method for an estimation of the 3D bounding box. The 3D bounding box is used to normalize the image viewpoint by "unpacking" the image into a plane. We also propose to randomly alter the color of the image and add a rectangle with random noise to a random position in the image during the training of Convolutional Neural Networks. We have collected a large fine-grained vehicle dataset BoxCars116k, with 116k images of vehicles from various viewpoints taken by numerous surveillance cameras. We performed a number of experiments which show that our proposed method significantly improves CNN classification accuracy (the accuracy is increased by up to 12 percentage points and the error is reduced by up to 50% compared to CNNs without the proposed modifications). We also show that our method outperforms state-of-the-art methods for fine-grained recognition.
BibTeX:
@ARTICLE{
   author = {Jakub Sochor and Jakub {\v{S}}pa{\v{n}}hel and
	Adam Herout},
   title = {BoxCars: Improving Fine-Grained Recognition of
	Vehicles using 3-D Bounding Boxes in Traffic
	Surveillance},
   pages = {97--108},
   journal = {IEEE Transactions on Intelligent Transportation Systems},
   volume = {2019},
   number = {1},
   year = {2018},
   ISSN = {1524-9050},
   doi = {10.1109/TITS.2018.2799228},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php.en.iso-8859-2?id=11617}
}

Your IPv4 address: 54.82.99.169
Switch to https

DNSSEC [dnssec]