Convolutional neural networks for computer vision

Build Status GitHub License Python Version

This repo is used to research convolutional networks for task of computer vision. For this purpose, the repo contains (re)implementations of various classification, segmentation, detection, and pose estimation models and scripts for training/evaluating/converting.

The following frameworks are used:

For each supported framework, there is a PIP-package containing pure models without auxiliary scripts. List of packages:

Currently, models are mostly implemented on Gluon and then ported to other frameworks. Some models are pretrained on ImageNet-1K, CIFAR-10/100, SVHN, CUB-200-2011, Pascal VOC2012, ADE20K, Cityscapes, and COCO datasets. All pretrained weights are loaded automatically during use. See examples of such automatic loading of weights in the corresponding sections of the documentation dedicated to a particular package:

Installation

To use training/evaluating scripts as well as all models, you need to clone the repository and install dependencies:

git clone git@github.com:osmr/imgclsmob.git
pip install -r requirements.txt

Table of implemented classification models

Some remarks:

Model Gluon PyTorch Chainer Keras TF TF2 Paper Repo Year
AlexNet A A A A A A link link 2012
ZFNet A A A A A A link - 2013
VGG A A A A A A link - 2014
BN-VGG A A A A A A link - 2015
BN-Inception A A A - - A link - 2015
ResNet ABCDE ABCDE ABCDE A A ABCDE link link 2015
PreResNet ABCD ABCD ABCD A A ABCD link link 2016
ResNeXt ABCD ABCD ABCD A A ABCD link link 2016
SENet A A A A A A link link 2017
SE-ResNet ABCDE ABCDE ABCDE A A ABCDE link link 2017
SE-PreResNet ABCD ABCD ABCD A A ABCD link link 2017
SE-ResNeXt A A A A A A link link 2017
IBN-ResNet A A - - - A link link 2018
IBN-ResNeXt A A - - - A link link 2018
IBN-DenseNet A A - - - A link link 2018
AirNet A A A - - A link link 2018
AirNeXt A A A - - A link link 2018
BAM-ResNet A A A - - A link link 2018
CBAM-ResNet A A A - - A link link 2018
ResAttNet a a a - - - link link 2017
SKNet a a a - - - link link 2019
DIA-ResNet aBCD aBCD aBCD - - - link link 2019
DIA-PreResNet aBCD aBCD aBCD - - - link link 2019
PyramidNet ABCD ABCD ABCD - - ABCD link link 2016
DiracNetV2 A A A - - A link link 2017
ShaResNet a a a - - - link link 2017
CRU-Net A - - - - - link link 2018
DenseNet ABCD ABCD ABCD A A ABCD link link 2016
CondenseNet A A A - - - link link 2017
SparseNet a a a - - - link link 2018
PeleeNet A A A - - A link link 2018
Oct-ResNet abcd a a - - - link - 2019
Res2Net a - - - - - link - 2019
WRN ABCD ABCD ABCD - - a link link 2016
WRN-1bit BCD BCD BCD - - - link link 2018
DRN-C A A A - - A link link 2017
DRN-D A A A - - A link link 2017
DPN A A A - - A link link 2017
DarkNet Ref A A A A A A link link -
DarkNet Tiny A A A A A A link link -
DarkNet-19 a a a a a a link link -
DarkNet-53 A A A A A A link link 2018
ChannelNet a a a - a - link link 2018
iSQRT-COV-ResNet a a - - - - link link 2017
RevNet - a - - - - link link 2017
i-RevNet A A A - - - link link 2018
BagNet A A A - - A link link 2019
DLA A A A - - A link link 2017
MSDNet a ab - - - - link link 2017
FishNet A A A - - - link link 2018
ESPNetv2 A A A - - - link link 2018
HRNet A A A - - A link link 2019
VoVNet A A A - - A link link 2019
SelecSLS A A A - - A link link 2019
HarDNet A A A - - A link link 2019
X-DenseNet aBCD aBCD aBCD - - - link link 2017
SqueezeNet A A A A A A link link 2016
SqueezeResNet A A A A A A link - 2016
SqueezeNext A A A A A A link link 2018
ShuffleNet A A A A A A link - 2017
ShuffleNetV2 A A A A A A link - 2018
MENet A A A A A A link link 2018
MobileNet AE AE AE A A AE link link 2017
FD-MobileNet A A A A A A link link 2018
MobileNetV2 A A A A A A link link 2018
MobileNetV3 A A A A - A link link 2019
IGCV3 A A A A A A link link 2018
GhostNet a a a - - a link link 2019
MnasNet A A A A A A link - 2018
DARTS A A A - - - link link 2018
ProxylessNAS AE AE AE - - AE link link 2018
FBNet-C A A A - - A link - 2018
Xception A A A - - A link link 2016
InceptionV3 A A A - - A link link 2015
InceptionV4 A A A - - A link link 2016
InceptionResNetV2 A A A - - A link link 2016
PolyNet A A A - - A link link 2016
NASNet-Large A A A - - A link link 2017
NASNet-Mobile A A A - - A link link 2017
PNASNet-Large A A A - - A link link 2017
SPNASNet A A A - - A link link 2019
EfficientNet A A A A - A link link 2019
MixNet A A A - - A link link 2019
NIN BCD BCD BCD - - - link link 2013
RoR-3 BCD BCD BCD - - - link - 2016
RiR BCD BCD BCD - - - link - 2016
ResDrop-ResNet bcd bcd bcd - - - link link 2016
Shake-Shake-ResNet BCD BCD BCD - - - link link 2017
ShakeDrop-ResNet bcd bcd bcd - - - link - 2018
FractalNet bc bc - - - - link link 2016
NTS-Net E E E - - - link link 2018

Table of implemented segmentation models

Some remarks:

Model Gluon PyTorch Chainer Keras TF TF2 Paper Repo Year
PSPNet ABCD ABCD ABCD - - ABCD link - 2016
DeepLabv3 ABcD ABcD ABcD - - ABcD link - 2017
FCN-8s(d) ABcD ABcD ABcD - - ABcD link - 2014
ICNet C C C - - C link link 2017
SINet C C C - - c link link 2019
BiSeNet e e e - - e link - 2018

Table of implemented object detection models

Some remarks:

Model Gluon PyTorch Chainer Keras TF TF2 Paper Repo Year
CenterNet a a a - - a link link 2019

Table of implemented human pose estimation models

Some remarks:

Model Gluon PyTorch Chainer Keras TF TF2 Paper Repo Year
AlphaPose A A A - - A link link 2016
SimplePose A A A - - A link link 2018
SimplePose(Mobile) A A A - - A link - 2018
Lightweight OpenPose A A A - - A link link 2018
IBPPose A A A - - A link link 2019