Image Classification with CondenseNeXt for ARM-Based Computing Platforms

TL;DR

This paper introduces CondenseNeXt, an ultra-efficient deep CNN architecture optimized for ARM-based embedded platforms, achieving state-of-the-art accuracy with significantly reduced computational requirements on multiple benchmark datasets.

Contribution

The paper presents CondenseNeXt, a novel CNN architecture that combines depthwise separable convolutions and model compression for high efficiency on ARM-based devices.

Findings

Achieves state-of-the-art accuracy on CIFAR-10, CIFAR-100, and ImageNet datasets.

Reduces FLOPs by up to 59.98% compared to CondenseNet.

Operates efficiently on ARM-based platforms without GPU support.

Abstract

In this paper, we demonstrate the implementation of our ultra-efficient deep convolutional neural network architecture: CondenseNeXt on NXP BlueBox, an autonomous driving development platform developed for self-driving vehicles. We show that CondenseNeXt is remarkably efficient in terms of FLOPs, designed for ARM-based embedded computing platforms with limited computational resources and can perform image classification without the need of a CUDA enabled GPU. CondenseNeXt utilizes the state-of-the-art depthwise separable convolution and model compression techniques to achieve a remarkable computational efficiency. Extensive analyses are conducted on CIFAR-10, CIFAR-100 and ImageNet datasets to verify the performance of CondenseNeXt Convolutional Neural Network (CNN) architecture. It achieves state-of-the-art image classification performance on three benchmark datasets including CIFAR-10 (4.79% top-1 error), CIFAR-100 (21.98% top-1 error) and ImageNet (7.91% single model, single crop top-5 error). CondenseNeXt achieves final trained model size improvement of 2.9+ MB and up to 59.98% reduction in forward FLOPs compared to CondenseNet and can perform image classification on ARM-Based computing platforms without needing a CUDA enabled GPU support, with outstanding efficiency.