All Projects → RangiLyu → mobilenext

RangiLyu / mobilenext

Licence: other
non-official PyTorch implementation of MobileNeXt from paper [Rethinking Bottleneck Structure for Efficient Mobile Network Design]

Programming Languages

python
139335 projects - #7 most used programming language

Projects that are alternatives of or similar to mobilenext

Sightseq
Computer vision tools for fairseq, containing PyTorch implementation of text recognition and object detection
Stars: ✭ 116 (+169.77%)
Mutual labels:  mobilenet
Posenet Coreml
I checked the performance by running PoseNet on CoreML
Stars: ✭ 143 (+232.56%)
Mutual labels:  mobilenet
Mobilenetv3 Tensorflow
Unofficial implementation of MobileNetV3 architecture described in paper Searching for MobileNetV3.
Stars: ✭ 185 (+330.23%)
Mutual labels:  mobilenet
Mobilenet
A Clearer and Simpler MobileNet Implementation in TensorFlow
Stars: ✭ 119 (+176.74%)
Mutual labels:  mobilenet
Paddlex
PaddlePaddle End-to-End Development Toolkit(『飞桨』深度学习全流程开发工具)
Stars: ✭ 3,399 (+7804.65%)
Mutual labels:  mobilenet
Caffe Mobilenet
A caffe implementation of mobilenet's depthwise convolution layer.
Stars: ✭ 146 (+239.53%)
Mutual labels:  mobilenet
Face landmark dnn
Face Landmark Detector based on Mobilenet V1
Stars: ✭ 92 (+113.95%)
Mutual labels:  mobilenet
Keras Semantic Segmentation
Keras-Semantic-Segmentation
Stars: ✭ 228 (+430.23%)
Mutual labels:  mobilenet
Mobilenet Ssd
Caffe implementation of Google MobileNet SSD detection network, with pretrained weights on VOC0712 and mAP=0.727.
Stars: ✭ 1,805 (+4097.67%)
Mutual labels:  mobilenet
Mobilenet Ssd Tensorrt
Accelerate mobileNet-ssd with tensorRT
Stars: ✭ 183 (+325.58%)
Mutual labels:  mobilenet
Object Localization
Object localization in images using simple CNNs and Keras
Stars: ✭ 130 (+202.33%)
Mutual labels:  mobilenet
Mk Tfjs
Play MK.js with TensorFlow.js
Stars: ✭ 133 (+209.3%)
Mutual labels:  mobilenet
Pytorch Cifar100
Practice on cifar100(ResNet, DenseNet, VGG, GoogleNet, InceptionV3, InceptionV4, Inception-ResNetv2, Xception, Resnet In Resnet, ResNext,ShuffleNet, ShuffleNetv2, MobileNet, MobileNetv2, SqueezeNet, NasNet, Residual Attention Network, SENet, WideResNet)
Stars: ✭ 2,423 (+5534.88%)
Mutual labels:  mobilenet
Faster Mobile Retinaface
[CVPR 2020] Reimplementation of RetinaFace, faster and stronger.
Stars: ✭ 117 (+172.09%)
Mutual labels:  mobilenet
Mobilenetv3 Ssd
MobileNetV3-SSD for object detection and implementation in PyTorch
Stars: ✭ 188 (+337.21%)
Mutual labels:  mobilenet
Mobilefacenet
Stars: ✭ 114 (+165.12%)
Mutual labels:  mobilenet
Efficientnet
Implementation of EfficientNet model. Keras and TensorFlow Keras.
Stars: ✭ 1,920 (+4365.12%)
Mutual labels:  mobilenet
Octconv.pytorch
PyTorch implementation of Octave Convolution with pre-trained Oct-ResNet and Oct-MobileNet models
Stars: ✭ 229 (+432.56%)
Mutual labels:  mobilenet
Swiftopenpose
tf-openpose Based iOS Project
Stars: ✭ 215 (+400%)
Mutual labels:  mobilenet
Proctoring Ai
Creating a software for automatic monitoring in online proctoring
Stars: ✭ 155 (+260.47%)
Mutual labels:  mobilenet

MobileNeXt: Rethinking Bottleneck Structure for Efficient Mobile Network Design

Introduction

This is a non-official PyTorch implementation of MobileNeXt model from this paper Rethinking Bottleneck Structure for Efficient Mobile Network Design

Details

Architecture

The following is the architecture details of MobileNeXt

No. t Out-Dim s b Inp-Dim Operater
1 - 112 × 112 × 32 2 1 224 × 224 × 3 conv2d 3x3
2 2 56 × 56 × 96 2 1 112 × 112 × 32 sandglass block
3 6 56 × 56 × 144 1 1 56 × 56 × 96 sandglass block
4 6 28 × 28 × 192 2 3 56 × 56 × 144 sandglass block
5 6 14 × 14 × 288 2 3 28 × 28 × 192 sandglass block
6 6 14 × 14 × 384 1 4 14 × 14 × 288 sandglass block
7 6 7 × 7 × 576 2 4 14 × 14 × 384 sandglass block
8 6 7 × 7 × 960 1 2 7 × 7 × 576 sandglass block
9 6 7 × 7 × 1280 1 1 7 × 7 × 960 sandglass block
10 - 1 × 1 × 1280 - 1 7 × 7 × 1280 avgpool 7x7
11 - k - 1 1 × 1 × 1280 conv2d 1x1

Sand Glass Module

sandglass_image

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].