Vehicle Fine-grained Recognition Based on Convolutional Neural Networks for Real-world Applications

Abstract

We explore the implementation of vehicle fine-grained type and color recognition based on neural networks in a real-world application. We suggest changes to the previously published method with respect to capabilities of low-powered devices, such as Nvidia Jetson. Experimental evaluation shows that the accuracy of MobileNet net slightly decreases compared to ResNet-50 from 89.55% to 86.13% while inference is 2.4× faster on Jetson.