A Large-scale and Global Car Dataset for Verification

DOI

10.2991/cset-16.2016.12How to use a DOI?

Keywords

Car model, dataset, Gcars, car verification

Abstract

Few researches focus on the larger-scale car model dataset compared with other objects in computer vision for verification, such as classification and face verification. In this paper, we present our on-going effort in collecting a large-scale and global dataset, Gcars, for improving related car model research. This dataset contains not only the most famous global car models but also the most locals in China, where all car images are collected from public website and the car hierarchy is a three-layer, make, model and type. We also demonstrate the most important application, car verification, exploiting the dataset. The performance in terms of verification accuracy is better than that of the benchmark dataset, namely CompCars, which is the similar and famous dataset, by using the deep learning framework Caffe.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Lingji Hu
AU  - Xingcheng Luo
AU  - Jianhua Deng
AU  - Fengjie Lai
AU  - Jian Hu
AU  - Yongbin Yu
PY  - 2016/08
DA  - 2016/08
TI  - A Large-scale and Global Car Dataset for Verification
BT  - Proceedings of the 2016 International Conference on Computer Science and Electronic Technology
PB  - Atlantis Press
SP  - 49
EP  - 52
SN  - 2352-538X
UR  - https://doi.org/10.2991/cset-16.2016.12
DO  - 10.2991/cset-16.2016.12
ID  - Hu2016/08
ER  -