Enabling Image Recognition on Constrained Devices Using Neural Network Pruning and a CycleGAN

TL;DR

This paper presents methods to enable efficient image recognition on constrained devices by applying neural network pruning to reduce model size and using CycleGANs to adapt out-of-distribution images, facilitating edge AI in smart cameras.

Contribution

It introduces a novel combination of neural network pruning and CycleGAN-based image transformation to improve resource efficiency for image recognition on IoT devices.

Findings

Neural network pruning retains high accuracy with only 1.1% of neurons.

CycleGAN effectively transforms out-of-distribution images to in-domain images.

Combined approaches enable practical edge AI deployment in smart cameras.

Abstract

Smart cameras are increasingly used in surveillance solutions in public spaces. Contemporary computer vision applications can be used to recognize events that require intervention by emergency services. Smart cameras can be mounted in locations where citizens feel particularly unsafe, e.g., pathways and underpasses with a history of incidents. One promising approach for smart cameras is edge AI, i.e., deploying AI technology on IoT devices. However, implementing resource-demanding technology such as image recognition using deep neural networks (DNN) on constrained devices is a substantial challenge. In this paper, we explore two approaches to reduce the need for compute in contemporary image recognition in an underpass. First, we showcase successful neural network pruning, i.e., we retain comparable classification accuracy with only 1.1\% of the neurons remaining from the state-of-the-art DNN architecture. Second, we demonstrate how a CycleGAN can be used to transform out-of-distribution images to the operational design domain. We posit that both pruning and CycleGANs are promising enablers for efficient edge AI in smart cameras.