A Large and Diverse Dataset for Improved Vehicle Make and Model Recognition

Faezeh Tafazzoli, Hichem Frigui, Keishin Nishiyama; Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2017, pp. 1-8

Abstract


Vehicle Make and Model Recognition (VMMR) has evolved into a significant subject of study due to its importance in numerous Intelligent Transportation Systems (ITS) and its components such as Automated Vehicular Surveillance (AVS). A highly accurate and real-time VMMR system significantly reduces the overhead cost of resources other-wise required. The VMMR problem is a multi-class classification task with a peculiar set of issues and challenges like multiplicity, inter- and intra-make ambiguity among various classes, which need to be solved in an efficient and reliable manner to achieve a highly robust VMMR system. In this paper, facing the growing importance of make and model recognition of vehicles, we present a dataset with 9170 different classes of vehicles to advance the corresponding tasks. Extensive experiments conducted using baseline approaches yield superior results for images that were occluded, under low illumination or partial camera views, available in our VMMR dataset.

Related Material


[pdf]
[bibtex]
@InProceedings{Tafazzoli_2017_CVPR_Workshops,
author = {Tafazzoli, Faezeh and Frigui, Hichem and Nishiyama, Keishin},
title = {A Large and Diverse Dataset for Improved Vehicle Make and Model Recognition},
booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {July},
year = {2017}
}