Car Type Recognition with Deep Neural Networks

TL;DR

This paper compares deep neural networks and SIFT-based support vector machines for automatic car type recognition, achieving over 97% accuracy with a large dataset, surpassing previous manual feature methods.

Contribution

It demonstrates the effectiveness of deep neural networks for car type recognition, significantly improving accuracy over traditional manual feature extraction approaches.

Findings

Deep neural networks achieve over 97% accuracy.

Support vector machines with SIFT features also perform well.

Deep learning outperforms manual feature methods.

Abstract

In this paper we study automatic recognition of cars of four types: Bus, Truck, Van and Small car. For this problem we consider two data driven frameworks: a deep neural network and a support vector machine using SIFT features. The accuracy of the methods is validated with a database of over 6500 images, and the resulting prediction accuracy is over 97 %. This clearly exceeds the accuracies of earlier studies that use manually engineered feature extraction pipelines.