CNN 网络整理 (持续更新)

基本网络

• AlexNet: Imagenet classification with deep convolutional neural networks, NIPS, 2012
• CaffeNet: Caffe: Convolutional Architecture for Fast Feature Embedding, ACMMM, 2014
• VGGNet: Very deep convolutional networks for large-scale image recognition, ICLR, 2015
• Inception-v1 [GoogLeNet]: Going deeper with convolutions, CVPR, 2015
• Inception-v2 [BN-Inception]: Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift, arXiv, 2015
• Inception-v3: Rethinking the Inception Architecture for Computer Vision, CVPR, 2016
• Inception-v4, Inception-ResNet-v1 : Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning, AAAI, 2017
• ResNet: Deep Residual Learning for Image Recognition, CVPR, 2016
• ResNet-v2: Identity Mappings in Deep Residual Networks, ECCV, 2016
• Inception-ResNet-v2: no paper
• WRN: Wide Residual Networks, arXiv, 2016
• ResNeXt: Aggregated residual transformations for deep neural networks, CVPR, 2017
• DiracNet: DiracNets: Training Very Deep Neural Networks Without Skip-Connections, arXiv, 2017
• FractalNet: Ultra-Deep Neural Networks without Residuals, ICLR, 2017
• DenseNet: Densely Connected Convolutional Networks, CVPR, 2017
• DPN, Dual Path Networks, NIPS, 2017
• SENet: Squeeze-and-Excitation Networks, CVPR, 2018
• IGC: Interleaved Group Convolutions, ICCV, 2017
• IGC-v2: IGCV2: Interleaved Structured Sparse Convolutional Neural Networks, CVPR, 2018
• IGC-v3: IGCV3: Interleaved Low-Rank Group Convolutions for Efficient Deep Neural Networks, arXiv, 2018
• NASNet: Learning Transferable Architectures for Scalable Image Recognition, CVPR, 2018

轻量级网络

• SqueezeNet: SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB, ICLR, 2017
• MobileNet: MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications, CVPR, 2017
• ShuffleNet: ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices, CVPR, 2017
• Xception: Xception: Deep Learning with Depth-wise Separable Convolutions, CVPR, 2017
• ShuffleNet-v2: ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design, arXiv, 2018
• MobileNet-v2: MobileNetV2: Inverted Residuals and Linear Bottlenecks, CVPR, 2018

理论

• An overview of gradient descent optimization algorithms, arXiv, 2016
• Deep Networks with Stochastic Depth, ECCV, 2016
• Residual Networks Behave Like Ensembles of Relatively Shallow Networks, NIPS, 2016

经验

• All you need is a good init, arXiv, 2015
• Recent Advances in Convolutional Neural Networks, arXiv, 2015
• A guide to convolution arithmetic for deep learning, arXiv, 2016
• An Analysis of Deep Neural Network Models for Practical Applications, arXiv, 2016