└── README.md /README.md: -------------------------------------------------------------------------------- 1 | 2 | 3 | # Awesome - Image Classification 4 | 5 | [![Awesome](https://cdn.rawgit.com/sindresorhus/awesome/d7305f38d29fed78fa85652e3a63e154dd8e8829/media/badge.svg)](https://github.com/sindresorhus/awesome) 6 | 7 | A curated list of deep learning image classification papers and codes since 2014, Inspired by [awesome-object-detection](https://github.com/amusi/awesome-object-detection), [deep_learning_object_detection](https://github.com/hoya012/deep_learning_object_detection) and [awesome-deep-learning-papers](https://github.com/terryum/awesome-deep-learning-papers). 8 | 9 | ## Background 10 | 11 | I believe image classification is a great start point before diving into other computer vision fields, espacially 12 | for begginers who know nothing about deep learning. When I started to learn computer vision, I've made a lot of mistakes, I wish someone could have told me that which paper I should start with back then. There doesn't seem to have a repository to have a list of image classification papers like [deep_learning_object_detection](https://github.com/hoya012/deep_learning_object_detection) until now. Therefore, I decided to make a repository 13 | of a list of deep learning image classification papers and codes to help others. My personal advice for people who 14 | know nothing about deep learning, try to start with vgg, then googlenet, resnet, feel free to continue reading other listed papers or switch to other fields after you are finished. 15 | 16 | **Note: I also have a repository of pytorch implementation of some of the image classification networks, you can check out [here](https://github.com/weiaicunzai/pytorch-cifar).** 17 | 18 | ## Performance Table 19 | 20 | For simplicity reason, I only listed the best top1 and top5 accuracy on ImageNet from the papers. Note that this does not necessarily mean one network is better than another when the acc is higher, cause some networks are focused on reducing the model complexity instead of improving accuracy, or some papers only give the single crop results on ImageNet, but others give the model fusion or multicrop results. 21 | 22 | - ConvNet: name of the covolution network 23 | - ImageNet top1 acc: best top1 accuracy on ImageNet from the Paper 24 | - ImageNet top5 acc: best top5 accuracy on ImageNet from the Paper 25 | - Published In: which conference or journal the paper was published in. 26 | 27 | | ConvNet | ImageNet top1 acc | ImageNet top5 acc | Published In | 28 | |:--------------------------:|:-----------------:|:-----------------:|:------------------:| 29 | | Vgg | 76.3 | 93.2 | ICLR2015 | 30 | | GoogleNet | - | 93.33 | CVPR2015 | 31 | | PReLU-nets | - | 95.06 | ICCV2015 | 32 | | ResNet | - | 96.43 | CVPR2015 | 33 | | PreActResNet | 79.9 | 95.2 | CVPR2016 | 34 | | Inceptionv3 | 82.8 | 96.42 | CVPR2016 | 35 | | Inceptionv4 | 82.3 | 96.2 | AAAI2016 | 36 | | Inception-ResNet-v2 | 82.4 | 96.3 | AAAI2016 | 37 | |Inceptionv4 + Inception-ResNet-v2| 83.5 | 96.92 | AAAI2016 | 38 | | RiR | - | - | ICLR Workshop2016 | 39 | | Stochastic Depth ResNet | 78.02 | - | ECCV2016 | 40 | | WRN | 78.1 | 94.21 | BMVC2016 | 41 | | SqueezeNet | 60.4 | 82.5 | arXiv2017([rejected by ICLR2017](https://openreview.net/forum?id=S1xh5sYgx)) | 42 | | GeNet | 72.13 | 90.26 | ICCV2017 | 43 | | MetaQNN | - | - | ICLR2017 | 44 | | PyramidNet | 80.8 | 95.3 | CVPR2017 | 45 | | DenseNet | 79.2 | 94.71 | ECCV2017 | 46 | | FractalNet | 75.8 | 92.61 | ICLR2017 | 47 | | ResNext | - | 96.97 | CVPR2017 | 48 | | IGCV1 | 73.05 | 91.08 | ICCV2017 | 49 | | Residual Attention Network | 80.5 | 95.2 | CVPR2017 | 50 | | Xception | 79 | 94.5 | CVPR2017 | 51 | | MobileNet | 70.6 | - | arXiv2017 | 52 | | PolyNet | 82.64 | 96.55 | CVPR2017 | 53 | | DPN | 79 | 94.5 | NIPS2017 | 54 | | Block-QNN | 77.4 | 93.54 | CVPR2018 | 55 | | CRU-Net | 79.7 | 94.7 | IJCAI2018 | 56 | | ShuffleNet | 75.3 | - | CVPR2018 | 57 | | CondenseNet | 73.8 | 91.7 | CVPR2018 | 58 | | NasNet | 82.7 | 96.2 | CVPR2018 | 59 | | MobileNetV2 | 74.7 | - | CVPR2018 | 60 | | IGCV2 | 70.07 | - | CVPR2018 | 61 | | hier | 79.7 | 94.8 | ICLR2018 | 62 | | PNasNet | 82.9 | 96.2 | ECCV2018 | 63 | | AmoebaNet | 83.9 | 96.6 | arXiv2018 | 64 | | SENet | - | 97.749 | CVPR2018 | 65 | | ShuffleNetV2 | 81.44 | - | ECCV2018 | 66 | | IGCV3 | 72.2 | - | BMVC2018 | 67 | | MnasNet | 76.13 | 92.85 | arXiv2018 | 68 | 69 | 70 | ## Papers&Codes 71 | 72 | ### VGG 73 | **Very Deep Convolutional Networks for Large-Scale Image Recognition.** 74 | Karen Simonyan, Andrew Zisserman 75 | - pdf: [https://arxiv.org/abs/1409.1556](https://arxiv.org/abs/1409.1556) 76 | - code: [torchvision : https://github.com/pytorch/vision/blob/master/torchvision/models/vgg.py](https://github.com/pytorch/vision/blob/master/torchvision/models/vgg.py) 77 | - code: [keras-applications : https://github.com/keras-team/keras-applications/blob/master/keras_applications/vgg16.py](https://github.com/keras-team/keras-applications/blob/master/keras_applications/vgg16.py) 78 | - code: [keras-applications : https://github.com/keras-team/keras-applications/blob/master/keras_applications/vgg19.py](https://github.com/keras-team/keras-applications/blob/master/keras_applications/vgg19.py) 79 | 80 | ### GoogleNet 81 | **Going Deeper with Convolutions** 82 | Christian Szegedy, Wei Liu, Yangqing Jia, Pierre Sermanet, Scott Reed, Dragomir Anguelov, Dumitru Erhan, Vincent Vanhoucke, Andrew Rabinovich 83 | - pdf: [https://arxiv.org/abs/1409.4842](https://arxiv.org/abs/1409.4842) 84 | - code: [unofficial-tensorflow : https://github.com/conan7882/GoogLeNet-Inception](https://github.com/conan7882/GoogLeNet-Inception) 85 | - code: [unofficial-caffe : https://github.com/lim0606/caffe-googlenet-bn](https://github.com/lim0606/caffe-googlenet-bn) 86 | 87 | ### PReLU-nets 88 | **Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification** 89 | Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun 90 | - pdf: [https://arxiv.org/abs/1502.01852](https://arxiv.org/abs/1502.01852) 91 | - code: [unofficial-chainer : https://github.com/nutszebra/prelu_net](https://github.com/nutszebra/prelu_net) 92 | 93 | ### ResNet 94 | **Deep Residual Learning for Image Recognition** 95 | Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun 96 | - pdf: [https://arxiv.org/abs/1512.03385](https://arxiv.org/abs/1512.03385) 97 | - code: [facebook-torch : https://github.com/facebook/fb.resnet.torch](https://github.com/facebook/fb.resnet.torch) 98 | - code: [torchvision : https://github.com/pytorch/vision/blob/master/torchvision/models/resnet.py](https://github.com/pytorch/vision/blob/master/torchvision/models/resnet.py) 99 | - code: [keras-applications : https://github.com/keras-team/keras-applications/blob/master/keras_applications/resnet.py](https://github.com/keras-team/keras-applications/blob/master/keras_applications/resnet.py) 100 | - code: [unofficial-keras : https://github.com/raghakot/keras-resnet](https://github.com/raghakot/keras-resnet) 101 | - code: [unofficial-tensorflow : https://github.com/ry/tensorflow-resnet](https://github.com/ry/tensorflow-resnet) 102 | 103 | ### PreActResNet 104 | **Identity Mappings in Deep Residual Networks** 105 | Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun 106 | - pdf: [https://arxiv.org/abs/1603.05027](https://arxiv.org/abs/1603.05027) 107 | - code: [facebook-torch : https://github.com/facebook/fb.resnet.torch/blob/master/models/preresnet.lua](https://github.com/facebook/fb.resnet.torch/blob/master/models/preresnet.lua) 108 | - code: [official : https://github.com/KaimingHe/resnet-1k-layers](https://github.com/KaimingHe/resnet-1k-layers) 109 | - code: [unoffical-pytorch : https://github.com/kuangliu/pytorch-cifar/blob/master/models/preact_resnet.py](https://github.com/kuangliu/pytorch-cifar/blob/master/models/preact_resnet.py) 110 | - code: [unoffical-mxnet : https://github.com/tornadomeet/ResNet](https://github.com/tornadomeet/ResNet) 111 | 112 | ### Inceptionv3 113 | **Rethinking the Inception Architecture for Computer Vision** 114 | Christian Szegedy, Vincent Vanhoucke, Sergey Ioffe, Jonathon Shlens, Zbigniew Wojna 115 | - pdf: [https://arxiv.org/abs/1512.00567](https://arxiv.org/abs/1512.00567) 116 | - code: [torchvision : https://github.com/pytorch/vision/blob/master/torchvision/models/inception.py](https://github.com/pytorch/vision/blob/master/torchvision/models/inception.py) 117 | - code: [keras-applications : https://github.com/keras-team/keras-applications/blob/master/keras_applications/inception_v3.py](https://github.com/keras-team/keras-applications/blob/master/keras_applications/inception_v3.py) 118 | 119 | ### Inceptionv4 && Inception-ResNetv2 120 | **Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning** 121 | Christian Szegedy, Sergey Ioffe, Vincent Vanhoucke, Alex Alemi 122 | - pdf: [https://arxiv.org/abs/1602.07261](https://arxiv.org/abs/1602.07261) 123 | - code: [unofficial-keras : https://github.com/kentsommer/keras-inceptionV4](https://github.com/kentsommer/keras-inceptionV4) 124 | - code: [unofficial-keras : https://github.com/titu1994/Inception-v4](https://github.com/titu1994/Inception-v4) 125 | - code: [unofficial-keras : https://github.com/yuyang-huang/keras-inception-resnet-v2](https://github.com/yuyang-huang/keras-inception-resnet-v2) 126 | 127 | ### RiR 128 | **Resnet in Resnet: Generalizing Residual Architectures** 129 | Sasha Targ, Diogo Almeida, Kevin Lyman 130 | - pdf: [https://arxiv.org/abs/1603.08029](https://arxiv.org/abs/1603.08029) 131 | - code: [unofficial-tensorflow : https://github.com/SunnerLi/RiR-Tensorflow](https://github.com/SunnerLi/RiR-Tensorflow) 132 | - code: [unofficial-chainer : https://github.com/nutszebra/resnet_in_resnet](https://github.com/nutszebra/resnet_in_resnet) 133 | 134 | ### Stochastic Depth ResNet 135 | **Deep Networks with Stochastic Depth** 136 | Gao Huang, Yu Sun, Zhuang Liu, Daniel Sedra, Kilian Weinberger 137 | - pdf: [https://arxiv.org/abs/1603.09382](https://arxiv.org/abs/1603.09382) 138 | - code: [unofficial-torch : https://github.com/yueatsprograms/Stochastic_Depth](https://github.com/yueatsprograms/Stochastic_Depth) 139 | - code: [unofficial-chainer : https://github.com/yasunorikudo/chainer-ResDrop](https://github.com/yasunorikudo/chainer-ResDrop) 140 | - code: [unofficial-keras : https://github.com/dblN/stochastic_depth_keras](https://github.com/dblN/stochastic_depth_keras) 141 | 142 | ### WRN 143 | **Wide Residual Networks** 144 | Sergey Zagoruyko, Nikos Komodakis 145 | - pdf: [https://arxiv.org/abs/1605.07146](https://arxiv.org/abs/1605.07146) 146 | - code: [official : https://github.com/szagoruyko/wide-residual-networks](https://github.com/szagoruyko/wide-residual-networks) 147 | - code: [unofficial-pytorch : https://github.com/xternalz/WideResNet-pytorch](https://github.com/xternalz/WideResNet-pytorch) 148 | - code: [unofficial-keras : https://github.com/asmith26/wide_resnets_keras](https://github.com/asmith26/wide_resnets_keras) 149 | - code: [unofficial-pytorch : https://github.com/meliketoy/wide-resnet.pytorch](https://github.com/meliketoy/wide-resnet.pytorch) 150 | 151 | ### SqueezeNet 152 | **SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size** 153 | Forrest N. Iandola, Song Han, Matthew W. Moskewicz, Khalid Ashraf, William J. Dally, Kurt Keutzer 154 | - pdf: [https://arxiv.org/abs/1602.07360](https://arxiv.org/abs/1602.07360) 155 | - code: [torchvision : https://github.com/pytorch/vision/blob/master/torchvision/models/squeezenet.py](https://github.com/pytorch/vision/blob/master/torchvision/models/squeezenet.py) 156 | - code: [unofficial-caffe : https://github.com/DeepScale/SqueezeNet](https://github.com/DeepScale/SqueezeNet) 157 | - code: [unofficial-keras : https://github.com/rcmalli/keras-squeezenet](https://github.com/rcmalli/keras-squeezenet) 158 | - code: [unofficial-caffe : https://github.com/songhan/SqueezeNet-Residual](https://github.com/songhan/SqueezeNet-Residual) 159 | 160 | ### GeNet 161 | **Genetic CNN** 162 | Lingxi Xie, Alan Yuille 163 | - pdf: [https://arxiv.org/abs/1703.01513](https://arxiv.org/abs/1703.01513) 164 | - code: [unofficial-tensorflow : https://github.com/aqibsaeed/Genetic-CNN](https://github.com/aqibsaeed/Genetic-CNN) 165 | 166 | ### MetaQNN 167 | **Designing Neural Network Architectures using Reinforcement Learning** 168 | Bowen Baker, Otkrist Gupta, Nikhil Naik, Ramesh Raskar 169 | - pdf: [https://arxiv.org/abs/1611.02167](https://arxiv.org/abs/1611.02167) 170 | - code: [official : https://github.com/bowenbaker/metaqnn](https://github.com/bowenbaker/metaqnn) 171 | 172 | ### PyramidNet 173 | **Deep Pyramidal Residual Networks** 174 | Dongyoon Han, Jiwhan Kim, Junmo Kim 175 | - pdf: [https://arxiv.org/abs/1610.02915](https://arxiv.org/abs/1610.02915) 176 | - code: [official : https://github.com/jhkim89/PyramidNet](https://github.com/jhkim89/PyramidNet) 177 | - code: [unofficial-pytorch : https://github.com/dyhan0920/PyramidNet-PyTorch](https://github.com/dyhan0920/PyramidNet-PyTorch) 178 | 179 | ### DenseNet 180 | **Densely Connected Convolutional Networks** 181 | Gao Huang, Zhuang Liu, Laurens van der Maaten, Kilian Q. Weinberger 182 | - pdf: [https://arxiv.org/abs/1608.06993](https://arxiv.org/abs/1608.06993) 183 | - code: [official : https://github.com/liuzhuang13/DenseNet](https://github.com/liuzhuang13/DenseNet) 184 | - code: [unofficial-keras : https://github.com/titu1994/DenseNet](https://github.com/titu1994/DenseNet) 185 | - code: [unofficial-caffe : https://github.com/shicai/DenseNet-Caffe](https://github.com/shicai/DenseNet-Caffe) 186 | - code: [unofficial-tensorflow : https://github.com/YixuanLi/densenet-tensorflow](https://github.com/YixuanLi/densenet-tensorflow) 187 | - code: [unofficial-pytorch : https://github.com/YixuanLi/densenet-tensorflow](https://github.com/YixuanLi/densenet-tensorflow) 188 | - code: [unofficial-pytorch : https://github.com/bamos/densenet.pytorch](https://github.com/bamos/densenet.pytorch) 189 | - code: [unofficial-keras : https://github.com/flyyufelix/DenseNet-Keras](https://github.com/flyyufelix/DenseNet-Keras) 190 | 191 | ### FractalNet 192 | **FractalNet: Ultra-Deep Neural Networks without Residuals** 193 | Gustav Larsson, Michael Maire, Gregory Shakhnarovich 194 | - pdf: [https://arxiv.org/abs/1605.07648](https://arxiv.org/abs/1605.07648) 195 | - code: [unofficial-caffe : https://github.com/gustavla/fractalnet](https://github.com/gustavla/fractalnet) 196 | - code: [unofficial-keras : https://github.com/snf/keras-fractalnet](https://github.com/snf/keras-fractalnet) 197 | - code: [unofficial-tensorflow : https://github.com/tensorpro/FractalNet](https://github.com/tensorpro/FractalNet) 198 | 199 | ### ResNext 200 | **Aggregated Residual Transformations for Deep Neural Networks** 201 | Saining Xie, Ross Girshick, Piotr Dollár, Zhuowen Tu, Kaiming He 202 | - pdf: [https://arxiv.org/abs/1611.05431](https://arxiv.org/abs/1611.05431) 203 | - code: [official : https://github.com/facebookresearch/ResNeXt](https://github.com/facebookresearch/ResNeXt) 204 | - code: [keras-applications : https://github.com/keras-team/keras-applications/blob/master/keras_applications/resnext.py](https://github.com/keras-team/keras-applications/blob/master/keras_applications/resnext.py) 205 | - code: [unofficial-pytorch : https://github.com/prlz77/ResNeXt.pytorch](https://github.com/prlz77/ResNeXt.pytorch) 206 | - code: [unofficial-keras : https://github.com/titu1994/Keras-ResNeXt](https://github.com/titu1994/Keras-ResNeXt) 207 | - code: [unofficial-tensorflow : https://github.com/taki0112/ResNeXt-Tensorflow](https://github.com/taki0112/ResNeXt-Tensorflow) 208 | - code: [unofficial-tensorflow : https://github.com/wenxinxu/ResNeXt-in-tensorflow](https://github.com/wenxinxu/ResNeXt-in-tensorflow) 209 | 210 | ### IGCV1 211 | **Interleaved Group Convolutions for Deep Neural Networks** 212 | Ting Zhang, Guo-Jun Qi, Bin Xiao, Jingdong Wang 213 | - pdf: [https://arxiv.org/abs/1707.02725](https://arxiv.org/abs/1707.02725) 214 | - code [official : https://github.com/hellozting/InterleavedGroupConvolutions](https://github.com/hellozting/InterleavedGroupConvolutions) 215 | 216 | ### Residual Attention Network 217 | **Residual Attention Network for Image Classification** 218 | Fei Wang, Mengqing Jiang, Chen Qian, Shuo Yang, Cheng Li, Honggang Zhang, Xiaogang Wang, Xiaoou Tang 219 | - pdf: [https://arxiv.org/abs/1704.06904](https://arxiv.org/abs/1704.06904) 220 | - code: [official : https://github.com/fwang91/residual-attention-network](https://github.com/fwang91/residual-attention-network) 221 | - code: [unofficial-pytorch : https://github.com/tengshaofeng/ResidualAttentionNetwork-pytorch](https://github.com/tengshaofeng/ResidualAttentionNetwork-pytorch) 222 | - code: [unofficial-gluon : https://github.com/PistonY/ResidualAttentionNetwork](https://github.com/PistonY/ResidualAttentionNetwork) 223 | - code: [unofficial-keras : https://github.com/koichiro11/residual-attention-network](https://github.com/koichiro11/residual-attention-network) 224 | 225 | ### Xception 226 | **Xception: Deep Learning with Depthwise Separable Convolutions** 227 | François Chollet 228 | - pdf: [https://arxiv.org/abs/1610.02357](https://arxiv.org/abs/1610.02357) 229 | - code: [unofficial-pytorch : https://github.com/jfzhang95/pytorch-deeplab-xception/blob/master/modeling/backbone/xception.py](https://github.com/jfzhang95/pytorch-deeplab-xception/blob/master/modeling/backbone/xception.py) 230 | - code: [unofficial-tensorflow : https://github.com/kwotsin/TensorFlow-Xception](https://github.com/kwotsin/TensorFlow-Xception) 231 | - code: [unofficial-caffe : https://github.com/yihui-he/Xception-caffe](https://github.com/yihui-he/Xception-caffe) 232 | - code: [unofficial-pytorch : https://github.com/tstandley/Xception-PyTorch](https://github.com/tstandley/Xception-PyTorch) 233 | - code: [keras-applications : https://github.com/keras-team/keras-applications/blob/master/keras_applications/xception.py](https://github.com/keras-team/keras-applications/blob/master/keras_applications/xception.py) 234 | 235 | ### MobileNet 236 | **MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications** 237 | Andrew G. Howard, Menglong Zhu, Bo Chen, Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, Hartwig Adam 238 | - pdf: [https://arxiv.org/abs/1704.04861](https://arxiv.org/abs/1704.04861) 239 | - code: [unofficial-tensorflow : https://github.com/Zehaos/MobileNet](https://github.com/Zehaos/MobileNet) 240 | - code: [unofficial-caffe : https://github.com/shicai/MobileNet-Caffe](https://github.com/shicai/MobileNet-Caffe) 241 | - code: [unofficial-pytorch : https://github.com/marvis/pytorch-mobilenet](https://github.com/marvis/pytorch-mobilenet) 242 | - code: [keras-applications : https://github.com/keras-team/keras-applications/blob/master/keras_applications/mobilenet.py](https://github.com/keras-team/keras-applications/blob/master/keras_applications/mobilenet.py) 243 | 244 | ### PolyNet 245 | **PolyNet: A Pursuit of Structural Diversity in Very Deep Networks** 246 | Xingcheng Zhang, Zhizhong Li, Chen Change Loy, Dahua Lin 247 | - pdf: [https://arxiv.org/abs/1611.05725](https://arxiv.org/abs/1611.05725) 248 | - code: [official : https://github.com/open-mmlab/polynet](https://github.com/open-mmlab/polynet) 249 | 250 | ### DPN 251 | **Dual Path Networks** 252 | Yunpeng Chen, Jianan Li, Huaxin Xiao, Xiaojie Jin, Shuicheng Yan, Jiashi Feng 253 | - pdf: [https://arxiv.org/abs/1707.01629](https://arxiv.org/abs/1707.01629) 254 | - code: [official : https://github.com/cypw/DPNs](https://github.com/cypw/DPNs) 255 | - code: [unoffical-keras : https://github.com/titu1994/Keras-DualPathNetworks](https://github.com/titu1994/Keras-DualPathNetworks) 256 | - code: [unofficial-pytorch : https://github.com/oyam/pytorch-DPNs](https://github.com/oyam/pytorch-DPNs) 257 | - code: [unofficial-pytorch : https://github.com/rwightman/pytorch-dpn-pretrained](https://github.com/rwightman/pytorch-dpn-pretrained) 258 | 259 | ### Block-QNN 260 | **Practical Block-wise Neural Network Architecture Generation** 261 | Zhao Zhong, Junjie Yan, Wei Wu, Jing Shao, Cheng-Lin Liu 262 | - pdf: [https://arxiv.org/abs/1708.05552](https://arxiv.org/abs/1708.05552) 263 | 264 | ### CRU-Net 265 | **Sharing Residual Units Through Collective Tensor Factorization in Deep Neural Networks** 266 | Chen Yunpeng, Jin Xiaojie, Kang Bingyi, Feng Jiashi, Yan Shuicheng 267 | - pdf[https://arxiv.org/abs/1703.02180](https://arxiv.org/abs/1703.02180) 268 | - code [official : https://github.com/cypw/CRU-Net](https://github.com/cypw/CRU-Net) 269 | - code [unofficial-mxnet : https://github.com/bruinxiong/Modified-CRUNet-and-Residual-Attention-Network.mxnet](https://github.com/bruinxiong/Modified-CRUNet-and-Residual-Attention-Network.mxnet) 270 | 271 | ### ShuffleNet 272 | **ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices** 273 | Xiangyu Zhang, Xinyu Zhou, Mengxiao Lin, Jian Sun 274 | - pdf: [https://arxiv.org/abs/1707.01083](https://arxiv.org/abs/1707.01083) 275 | - code: [unofficial-tensorflow : https://github.com/MG2033/ShuffleNet](https://github.com/MG2033/ShuffleNet) 276 | - code: [unofficial-pytorch : https://github.com/jaxony/ShuffleNet](https://github.com/jaxony/ShuffleNet) 277 | - code: [unofficial-caffe : https://github.com/farmingyard/ShuffleNet](https://github.com/farmingyard/ShuffleNet) 278 | - code: [unofficial-keras : https://github.com/scheckmedia/keras-shufflenet](https://github.com/scheckmedia/keras-shufflenet) 279 | 280 | ### CondenseNet 281 | **CondenseNet: An Efficient DenseNet using Learned Group Convolutions** 282 | Gao Huang, Shichen Liu, Laurens van der Maaten, Kilian Q. Weinberger 283 | - pdf: [https://arxiv.org/abs/1711.09224](https://arxiv.org/abs/1711.09224) 284 | - code: [official : https://github.com/ShichenLiu/CondenseNet](https://github.com/ShichenLiu/CondenseNet) 285 | - code: [unofficial-tensorflow : https://github.com/markdtw/condensenet-tensorflow](https://github.com/markdtw/condensenet-tensorflow) 286 | 287 | ### NasNet 288 | **Learning Transferable Architectures for Scalable Image Recognition** 289 | Barret Zoph, Vijay Vasudevan, Jonathon Shlens, Quoc V. Le 290 | - pdf: [https://arxiv.org/abs/1707.07012](https://arxiv.org/abs/1707.07012) 291 | - code: [unofficial-keras : https://github.com/titu1994/Keras-NASNet](https://github.com/titu1994/Keras-NASNet) 292 | - code: [keras-applications : https://github.com/keras-team/keras-applications/blob/master/keras_applications/nasnet.py](https://github.com/keras-team/keras-applications/blob/master/keras_applications/nasnet.py) 293 | - code: [unofficial-pytorch : https://github.com/wandering007/nasnet-pytorch](https://github.com/wandering007/nasnet-pytorch) 294 | - code: [unofficial-tensorflow : https://github.com/yeephycho/nasnet-tensorflow](https://github.com/yeephycho/nasnet-tensorflow) 295 | 296 | ### MobileNetV2 297 | **MobileNetV2: Inverted Residuals and Linear Bottlenecks** 298 | Mark Sandler, Andrew Howard, Menglong Zhu, Andrey Zhmoginov, Liang-Chieh Chen 299 | - pdf: [https://arxiv.org/abs/1801.04381](https://arxiv.org/abs/1801.04381) 300 | - code: [unofficial-keras : https://github.com/xiaochus/MobileNetV2](https://github.com/xiaochus/MobileNetV2) 301 | - code: [unofficial-pytorch : https://github.com/Randl/MobileNetV2-pytorch](https://github.com/Randl/MobileNetV2-pytorch) 302 | - code: [unofficial-tensorflow : https://github.com/neuleaf/MobileNetV2](https://github.com/neuleaf/MobileNetV2) 303 | 304 | ### IGCV2 305 | **IGCV2: Interleaved Structured Sparse Convolutional Neural Networks** 306 | Guotian Xie, Jingdong Wang, Ting Zhang, Jianhuang Lai, Richang Hong, Guo-Jun Qi 307 | - pdf: [https://arxiv.org/abs/1804.06202](https://arxiv.org/abs/1804.06202) 308 | 309 | ### hier 310 | **Hierarchical Representations for Efficient Architecture Search** 311 | Hanxiao Liu, Karen Simonyan, Oriol Vinyals, Chrisantha Fernando, Koray Kavukcuoglu 312 | - pdf: [https://arxiv.org/abs/1711.00436](https://arxiv.org/abs/1711.00436) 313 | 314 | ### PNasNet 315 | **Progressive Neural Architecture Search** 316 | Chenxi Liu, Barret Zoph, Maxim Neumann, Jonathon Shlens, Wei Hua, Li-Jia Li, Li Fei-Fei, Alan Yuille, Jonathan Huang, Kevin Murphy 317 | - pdf: [https://arxiv.org/abs/1712.00559](https://arxiv.org/abs/1712.00559) 318 | - code: [tensorflow-slim : https://github.com/tensorflow/models/blob/master/research/slim/nets/nasnet/pnasnet.py](https://github.com/tensorflow/models/blob/master/research/slim/nets/nasnet/pnasnet.py) 319 | - code: [unofficial-pytorch : https://github.com/chenxi116/PNASNet.pytorch](https://github.com/chenxi116/PNASNet.pytorch) 320 | - code: [unofficial-tensorflow : https://github.com/chenxi116/PNASNet.TF](https://github.com/chenxi116/PNASNet.TF) 321 | 322 | ### AmoebaNet 323 | **Regularized Evolution for Image Classifier Architecture Search** 324 | Esteban Real, Alok Aggarwal, Yanping Huang, Quoc V Le 325 | - pdf: [https://arxiv.org/abs/1802.01548](https://arxiv.org/abs/1802.01548) 326 | - code: [tensorflow-tpu : https://github.com/tensorflow/tpu/tree/master/models/official/amoeba_net](https://github.com/tensorflow/tpu/tree/master/models/official/amoeba_net) 327 | 328 | ### SENet 329 | **Squeeze-and-Excitation Networks** 330 | Jie Hu, Li Shen, Samuel Albanie, Gang Sun, Enhua Wu 331 | - pdf: [https://arxiv.org/abs/1709.01507](https://arxiv.org/abs/1709.01507) 332 | - code: [official : https://github.com/hujie-frank/SENet](https://github.com/hujie-frank/SENet) 333 | - code: [unofficial-pytorch : https://github.com/moskomule/senet.pytorch](https://github.com/moskomule/senet.pytorch) 334 | - code: [unofficial-tensorflow : https://github.com/taki0112/SENet-Tensorflow](https://github.com/taki0112/SENet-Tensorflow) 335 | - code: [unofficial-caffe : https://github.com/shicai/SENet-Caffe](https://github.com/shicai/SENet-Caffe) 336 | - code: [unofficial-mxnet : https://github.com/bruinxiong/SENet.mxnet](https://github.com/bruinxiong/SENet.mxnet) 337 | 338 | ### ShuffleNetV2 339 | **ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design** 340 | Ningning Ma, Xiangyu Zhang, Hai-Tao Zheng, Jian Sun 341 | - pdf: [https://arxiv.org/abs/1807.11164](https://arxiv.org/abs/1807.11164) 342 | - code: [unofficial-pytorch : https://github.com/Randl/ShuffleNetV2-pytorch](https://github.com/Randl/ShuffleNetV2-pytorch) 343 | - code: [unofficial-keras : https://github.com/opconty/keras-shufflenetV2](https://github.com/opconty/keras-shufflenetV2) 344 | - code: [unofficial-pytorch : https://github.com/Bugdragon/ShuffleNet_v2_PyTorch](https://github.com/Bugdragon/ShuffleNet_v2_PyTorch) 345 | - code: [unofficial-caff2: https://github.com/wolegechu/ShuffleNetV2.Caffe2](https://github.com/wolegechu/ShuffleNetV2.Caffe2) 346 | 347 | ### IGCV3 348 | **IGCV3: Interleaved Low-Rank Group Convolutions for Efficient Deep Neural Networks** 349 | Ke Sun, Mingjie Li, Dong Liu, Jingdong Wang 350 | - pdf: [https://arxiv.org/abs/1806.00178](https://arxiv.org/abs/1806.00178) 351 | - code: [official : https://github.com/homles11/IGCV3](https://github.com/homles11/IGCV3) 352 | - code: [unofficial-pytorch : https://github.com/xxradon/IGCV3-pytorch](https://github.com/xxradon/IGCV3-pytorch) 353 | - code: [unofficial-tensorflow : https://github.com/ZHANG-SHI-CHANG/IGCV3](https://github.com/ZHANG-SHI-CHANG/IGCV3) 354 | 355 | ### MNasNet 356 | **MnasNet: Platform-Aware Neural Architecture Search for Mobile** 357 | Mingxing Tan, Bo Chen, Ruoming Pang, Vijay Vasudevan, Quoc V. Le 358 | - pdf: [https://arxiv.org/abs/1807.11626](https://arxiv.org/abs/1807.11626) 359 | - code: [unofficial-pytorch : https://github.com/AnjieZheng/MnasNet-PyTorch](https://github.com/AnjieZheng/MnasNet-PyTorch) 360 | - code: [unofficial-caffe : https://github.com/LiJianfei06/MnasNet-caffe](https://github.com/LiJianfei06/MnasNet-caffe) 361 | - code: [unofficial-MxNet : https://github.com/chinakook/Mnasnet.MXNet](https://github.com/chinakook/Mnasnet.MXNet) 362 | - code: [unofficial-keras : https://github.com/Shathe/MNasNet-Keras-Tensorflow](https://github.com/Shathe/MNasNet-Keras-Tensorflow) 363 | 364 | --------------------------------------------------------------------------------