Convolutional Neural Network Optimization for Homography Estimation

Date of Award

2018

Degree Name

M.S. in Electrical Engineering

Department

Department of Electrical and Computer Engineering

Advisor/Chair

Advisor: Eric Balster

Abstract

This thesis proposes an optimized convolutional neural network architecture to improve homography estimation applications. The parameters and structure of the CNN including the number of convolutional filters, stride lengths, kernel size, learning parameters, etc are altered from previous implementations. Multiple modifications of the network are trained and evaluated until a final network yields a corner pixel error of 4.7 which is less than a network proposed in previous literature's.

Keywords

Electrical Engineering, Computer Engineering, Homography Estimation, Convolutional Neural Networks, Neural Networks, Image Registration, Deep Learning, Hyper-parameter Optimization

Rights Statement

Copyright © 2018, author

Link to Full Text

Share

COinS