Investigating Optimal Training and Uncertainty Quantification for CNN-based Optical Flow

Authors

  • Daiki Kurinara University of Notre Dame, Department of Aerospace and Mechanical Engineering, United States of America
  • Gianluca Blois University of Notre Dame, Department of Aerospace and Mechanical Engineering, United States of America
  • Hirotaka Sakaue University of Notre Dame, Department of Aerospace and Mechanical Engineering, United States of America
  • Daniele Schiavazzi University of Notre Dame, Department of Applied and Computational Mathematics and Statistics, United States of America

DOI:

https://doi.org/10.18409/ispiv.v1i1.131

Keywords:

Optical Flow, Machine Learning

Abstract

Optical Flow (OF) techniques provide “dense estimation” flow maps (i.e. pixel-level resolution) of timecorrelated images and thus are appealing to applications requiring high spatial resolutions. OF methods revolve around mathematical descriptions of the image as a collection of features, in which the pixel-level light intensity is the primary variable (Horn and Schunck, 1981). Feature tracking often involves the notion of scale invariance. Traditional OF approaches, merely based on mathematical formulations, have suffered from many challenges, especially when directly applied to images of fluid flows textured with tracer particles (hereafter PIV-like images). Due to the limited number of computationally manageable features and suboptimal regularization methods, successful implementation of past approaches has been limited to highly textured images and small displacement dynamic ranges.

Author Biography

Daiki Kurinara, University of Notre Dame, Department of Aerospace and Mechanical Engineering, United States of America

Daiki Kurinara is a PhD student at the Department of Aerospace and Mechanical Engineering at the University of Notre Dame.

Downloads

Published

2021-08-01

Issue

Section

Deep Learning and Data Assimilation