An Improved One millisecond Mobile Backbone

Efficient neural network backbones for mobile devices are often optimized for metrics such as FLOPs or parameter count. However, these metrics may not correlate well with latency of the network when deployed on a mobile device. Therefore, we perform extensive analysis of different metrics by deploying several mobile-friendly networks on a mobile device. We identify and analyze architectural and optimization bottlenecks in recent efficient neural networks and provide ways to mitigate these bottlenecks. To this end, we design an efficient backbone MobileOne, with variants achieving an inference time under 1 ms on an iPhone12 with 75.9 achieves state-of-the-art performance within the efficient architectures while being many times faster on mobile. Our best model obtains similar performance on ImageNet as MobileFormer while being 38x faster. Our model obtains 2.3 better top-1 accuracy on ImageNet than EfficientNet at similar latency. Furthermore, we show that our model generalizes to multiple tasks - image classification, object detection, and semantic segmentation with significant improvements in latency and accuracy as compared to existing efficient architectures when deployed on a mobile device.

READ FULL TEXT

page 16

page 17

research
03/24/2023

FastViT: A Fast Hybrid Vision Transformer using Structural Reparameterization

The recent amalgamation of transformer and convolutional designs has led...
research
11/09/2022

MFDNet: Towards Real-time Image Denoising On Mobile Devices

Deep convolutional neural networks have achieved great progress in image...
research
06/26/2020

Making DensePose fast and light

DensePose estimation task is a significant step forward for enhancing us...
research
12/24/2021

Virtuoso: Video-based Intelligence for real-time tuning on SOCs

Efficient and adaptive computer vision systems have been proposed to mak...
research
04/09/2018

NetAdapt: Platform-Aware Neural Network Adaptation for Mobile Applications

This work proposes an automated algorithm, called NetAdapt, that adapts ...
research
10/06/2021

ParaDiS: Parallelly Distributable Slimmable Neural Networks

When several limited power devices are available, one of the most effici...
research
04/11/2023

PP-MobileSeg: Explore the Fast and Accurate Semantic Segmentation Model on Mobile Devices

The success of transformers in computer vision has led to several attemp...

Please sign up or login with your details

Forgot password? Click here to reset