Convolutional Neural Networks for Aerial Vehicle Detection and Recognition

TL;DR

This paper presents a text-guided deep convolutional neural network for aerial vehicle recognition, capable of matching images with textual class descriptions, trained on synthetic data to handle more classes during testing.

Contribution

It introduces a novel text-guided CNN approach for aerial vehicle recognition, enabling flexible class matching beyond training classes.

Findings

Effective recognition of vehicles with class and color descriptions

Model generalizes to unseen class combinations

Demonstrates potential for scalable aerial vehicle classification

Abstract

This paper investigates the problem of aerial vehicle recognition using a text-guided deep convolutional neural network classifier. The network receives an aerial image and a desired class, and makes a yes or no output by matching the image and the textual description of the desired class. We train and test our model on a synthetic aerial dataset and our desired classes consist of the combination of the class types and colors of the vehicles. This strategy helps when considering more classes in testing than in training.