Visualizing and interpreting a deep neural network for classification of vehicular orientation

Permalink

https://hdl.handle.net/2142/99989

Title: Visualizing and interpreting a deep neural network for classification of vehicular orientation
Author(s): Du, Jianlin
Contributor(s): Varshney, Lav R.
Issue Date: 2018-05
Keyword(s): Deep learning
Abstract: We trained a deep neural network to classify images of cars facing 36 different directions, on a 2D image dataset rendered from 3D car models. After achieving a validation accuracy of 98.23%, we applied a series of interpretation techniques, including semantic dictionary, spatial attribution, and channel attribution, to the trained model, which enable us gain important insights on how the model recognized a car’s direction. For example, the channel attribution technique reveals in a certain layer, which filters contribute the most to distinguish a front facing car from a right facing car. Moreover, there are interactive interfaces for all the experiments, and readers could explore the interpretation of the model in a notebook environment.
Type of Resource: text
Language: en
Permalink: http://hdl.handle.net/2142/99989

The best of ECE undergraduate research