Deep 2.5D Vehicle Classification with Sparse SfM Depth Prior for Automated Toll Systems

TL;DR

This paper introduces a novel deep learning approach for vehicle classification in toll systems that leverages sparse 3D reconstruction as a regularizer, improving accuracy while only requiring 2D images at test time.

Contribution

The method uses sparse SfM-based 3D information as a regularizer in CNN training, enabling accurate vehicle classification with only 2D images during testing.

Findings

Improved classification accuracy over 2D methods.

Effective use of sparse 3D prior as a regularizer.

Test-time efficiency with only 2D images required.

Abstract

Automated toll systems rely on proper classification of the passing vehicles. This is especially difficult when the images used for classification only cover parts of the vehicle. To obtain information about the whole vehicle. we reconstruct the vehicle as 3D object and exploit this additional information within a Convolutional Neural Network (CNN). However, when using deep networks for 3D object classification, large amounts of dense 3D models are required for good accuracy, which are often neither available nor feasible to process due to memory requirements. Therefore, in our method we reproject the 3D object onto the image plane using the reconstructed points, lines or both. We utilize this sparse depth prior within an auxiliary network branch that acts as a regularizer during training. We show that this auxiliary regularizer helps to improve accuracy compared to 2D classification on a real-world dataset. Furthermore due to the design of the network, at test time only the 2D camera images are required for classification which enables the usage in portable computer vision systems.