Light-weight neural networks

Edit

Computer Vision • 19 methods

Lightweight neural networks or mobile-friendly neural networks are deep neural networks which are designed for mobile devices or other resource-limited hardwares

Methods

Add a Method

Method	Year	Papers
MobileNetV2 MobileNetV2: Inverted Residuals and Linear Bottlenecks	2018	321
SqueezeNet SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size	2016	85
MobileNetV3 Searching for MobileNetV3	2019	69
MobileNetV1 MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications	2017	66
ShuffleNet ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices	2017	48
MnasNet MnasNet: Platform-Aware Neural Architecture Search for Mobile	2018	20
GhostNet GhostNet: More Features from Cheap Operations	2019	18
ESPNet ESPNet: Efficient Spatial Pyramid of Dilated Convolutions for Semantic Segmentation	2018	18
FBNet FBNet: Hardware-Aware Efficient ConvNet Design via Differentiable Neural Architecture Search	2018	12
MobileViT MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformer	2021	10
MixNet MixConv: Mixed Depthwise Convolutional Kernels	2019	9
PeleeNet Pelee: A Real-Time Object Detection System on Mobile Devices	2018	3
MoViNet MoViNets: Mobile Video Networks for Efficient Video Recognition	2021	2
MobileDet MobileDets: Searching for Object Detection Architectures for Mobile Accelerators	2020	2
ProxylessNet-Mobile ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware	2018	2
MoGA-A MoGA: Searching Beyond MobileNetV3	2019	1
MoGA-B MoGA: Searching Beyond MobileNetV3	2019	1
MoGA-C MoGA: Searching Beyond MobileNetV3	2019	1
ESPNetv2 ESPNetv2: A Light-weight, Power Efficient, and General Purpose Convolutional Neural Network	2018	1