14 Cars Identification


A practical computer vision task is to identify the make and model of a motor vehicle within a photo. The MLHub package cars demonstrates the utility of Computer Vision for the identification of motor vehicles. We use the Comprehensive Cars dataset available for research purposes (Yang et al. 2015). The pre-built model was trained by Jiajun Zha through the Australian National University’s Software Innovation Institute. The pre-built model build code is available from github.

To install and configure the package:

ml install cars
ml configure cars
ml readme cars
ml commands cars

Note that on the first configure a 188MB file is downloaded, which is the pre-built model. This can take some time. It is cached locally so any further configuration simply uses the cached copy.

The package supports a identify command to identify the object (motor vehicle) in an image. The train command will use the tagged images within a folder of tagged to extend the pre-built model with new motor vehicle makes and models.

The source code for this MLHub package is available from gitlab: https://gitlab.com/kayontoga/cars.


Yang, Linjie, Ping Luo, Chen Change Loy, and Xiaoou Tang. 2015. β€œA Large-Scale Car Dataset for Fine-Grained Categorization and Verification.” Computer Vision and Pattern Recognition.

Your donation will support ongoing availability and give you access to the PDF version of this book. Desktop Survival Guides include Data Science, GNU/Linux, and MLHub. Books available on Amazon include Data Mining with Rattle and Essentials of Data Science. Popular open source software includes rattle, wajig, and mlhub. Hosted by Togaware, a pioneer of free and open source software since 1984. Copyright © 1995-2022 Graham.Williams@togaware.com Creative Commons Attribution-ShareAlike 4.0