BAR: Bayesian Activity Recognition using variational inference

TL;DR

This paper introduces a Bayesian deep learning approach for activity recognition that provides reliable uncertainty estimates and improves recognition performance, enhancing trustworthiness in safety-critical applications.

Contribution

It applies Bayesian inference with variational techniques to activity recognition DNNs, offering uncertainty quantification and improved accuracy over traditional methods.

Findings

Bayesian DNNs yield more reliable confidence measures.

The method improves recognition precision-recall AUC by 6.2%.

Models are evaluated on the Moments-In-Time dataset with in- and out-of-distribution samples.

Abstract

Uncertainty estimation in deep neural networks is essential for designing reliable and robust AI systems. Applications such as video surveillance for identifying suspicious activities are designed with deep neural networks (DNNs), but DNNs do not provide uncertainty estimates. Capturing reliable uncertainty estimates in safety and security critical applications will help to establish trust in the AI system. Our contribution is to apply Bayesian deep learning framework to visual activity recognition application and quantify model uncertainty along with principled confidence. We utilize the stochastic variational inference technique while training the Bayesian DNNs to infer the approximate posterior distribution around model parameters and perform Monte Carlo sampling on the posterior of model parameters to obtain the predictive distribution. We show that the Bayesian inference applied to DNNs provide reliable confidence measures for visual activity recognition task as compared to conventional DNNs. We also show that our method improves the visual activity recognition precision-recall AUC by 6.2% compared to non-Bayesian baseline. We evaluate our models on Moments-In-Time (MiT) activity recognition dataset by selecting a subset of in- and out-of-distribution video samples.