Please use this identifier to cite or link to this item:

Bayesian Augmentation of Deep Learning to Improve Video Classification

File Size Format  
0209.pdf 709.02 kB Adobe PDF View/Open

Item Summary

Title:Bayesian Augmentation of Deep Learning to Improve Video Classification
Authors:Swize, Emmie
Champagne, Lance
Cox, Bruce
Bihl, Trevor
Keywords:Soft Computing: Theory Innovations and Problem Solving Benefits
deep learning
show 1 morevideo surveillance
show less
Date Issued:04 Jan 2022
Abstract:Traditional automated video classification methods lack measures of uncertainty, meaning the network is unable to identify those cases in which its predictions are made with significant uncertainty. This leads to misclassification, as the traditional network classifies each observation with same amount of certainty, no matter what the observation is. Bayesian neural networks are a remedy to this issue by leveraging Bayesian inference to construct uncertainty measures for each prediction. Because exact Bayesian inference is typically intractable due to the large number of parameters in a neural network, Bayesian inference is approximated by utilizing dropout in a convolutional neural network. This research compared a traditional video classification neural network to its Bayesian equivalent based on performance and capabilities. The Bayesian network achieves higher accuracy than a comparable non-Bayesian video network and it further provides uncertainty measures for each classification.
Pages/Duration:10 pages
Rights:Attribution-NonCommercial-NoDerivatives 4.0 International
Appears in Collections: Soft Computing: Theory Innovations and Problem Solving Benefits

Please email if you need this content in ADA-compliant format.

This item is licensed under a Creative Commons License Creative Commons