Dimensionality of datasets in object detection networks

TL;DR

This paper investigates how the intrinsic dimensionality of data within different layers of CNNs affects object detection accuracy, especially when using augmented datasets, revealing differences in feature representations.

Contribution

It introduces an analysis of the intrinsic dimension's impact on CNNs for object detection and highlights differences between normal and augmented data representations.

Findings

Intrinsic dimension varies across layers.

Augmented data alters feature representations.

Representation differences affect detection accuracy.

Abstract

In recent years, convolutional neural networks (CNNs) are used in a large number of tasks in computer vision. One of them is object detection for autonomous driving. Although CNNs are used widely in many areas, what happens inside the network is still unexplained on many levels. Our goal is to determine the effect of Intrinsic dimension (i.e. minimum number of parameters required to represent data) in different layers on the accuracy of object detection network for augmented data sets. Our investigation determines that there is difference between the representation of normal and augmented data during feature extraction.