Dimensionality of datasets in object detection networks
This work addresses the problem of understanding internal CNN representations for autonomous driving object detection, but it is incremental as it focuses on a specific aspect without broad SOTA impact.
The study investigated how the intrinsic dimension of data in different layers affects object detection accuracy for augmented datasets in CNNs, finding differences in representation between normal and augmented data during feature extraction.
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.