Image Classification with CondenseNeXt for ARM-Based Computing Platforms

06/26/2021
by   Priyank Kalgaonkar, et al.
0

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 (21.98 error). CondenseNeXt achieves final trained model size improvement of 2.9+ MB and up to 59.98 perform image classification on ARM-Based computing platforms without needing a CUDA enabled GPU support, with outstanding efficiency.

READ FULL TEXT

page 1

page 2

page 5

research
11/28/2021

EffCNet: An Efficient CondenseNet for Image Classification on NXP BlueBox

Intelligent edge devices with built-in processors vary widely in terms o...
research
12/01/2021

CondenseNeXt: An Ultra-Efficient Deep Neural Network for Embedded Systems

Due to the advent of modern embedded systems and mobile devices with con...
research
05/28/2019

CompactNet: Platform-Aware Automatic Optimization for Convolutional Neural Networks

Convolutional Neural Network (CNN) based Deep Learning (DL) has achieved...
research
08/24/2019

Plexus Convolutional Neural Network (PlexusNet): A novel neural network architecture for histologic image analysis

Different convolutional neural network (CNN) models have been tested for...
research
01/26/2023

Efficient Hyperdimensional Computing

Hyperdimensional computing (HDC) uses binary vectors of high dimensions ...
research
11/12/2020

Gaussian RAM: Lightweight Image Classification via Stochastic Retina-Inspired Glimpse and Reinforcement Learning

Previous studies on image classification have mainly focused on the perf...
research
12/19/2017

DeepPicar: A Low-cost Deep Neural Network-based Autonomous Car

We present DeepPicar, a low-cost deep neural network based autonomous ca...

Please sign up or login with your details

Forgot password? Click here to reset