HCIILAB/CNN-Network-Architectures

CNN-Network-Architectures

popular network architecture list

• [Models]

---

AlexNet

VGG

GoogleNet

• Going deeper with convolutions

VGG2

• VGGFace2: A dataset for recognising faces across pose and age
• [vgg2]

---

ResNet

• Deep Residual Learning for Image Recognition
• [ResNet18]
• [ResNet50/101/152]

DenseNet

• Densely Connected Convolutional Networks
• [DenseNet121/161]

InceptionResNet v1, v2 / Inception v3, v4

• Rethinking the Inception Architecture for Computer Vision
• Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning
• [Inception v4]

ResNeXt-101

• Aggregated Residual Transformations for Deep Neural Networks
• [ResNeXt]

PyramidNet

• Deep Pyramidal Residual Networks
• [PyramidNet]

Dual Path Networks

• Dual Path Networks
• [DPNs]

NASNet

• Learning Transferable Architectures for Scalable Image Recognition

SENet

• Squeeze-and-Excitation Networks
• [SE-block]

InsightFace Net

• [LRestNetE-IR]

SKNet

• Selective Kernel Networks
• [SK-block]

Res2Net

• Res2Net: A New Multi-scale Backbone Architecture

EfficientNet

• EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks

IdleBlock

• Hybrid Composition with IdleBlock: More Efficient Networks for Image Recognition

---

Lightweight

SqueezeNet

• squeezenet alexnet-level accuracy with 50x fewer parameters and 0.5mb model size
• [SqueezeNet v1.0/v1.1]

MobileNet v1, v2

• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• MobileNetV2: Inverted Residuals and Linear Bottlenecks
• [MobileNet v1/v2]

ShuffleNet

• ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
• [ShuffleNet]
• [ShuffleNet]

SqueezeNext

• SqueezeNext: Hardware-Aware Neural Network Design
• [SqueezeNext]

Xception

• Xception: Deep Learning with Depthwise Separable Convolutions

MobileFaceNet

• MobileFaceNets: Efficient CNNs for Accurate Real-Time Face Verification on Mobile Devices
• [MobileFaceNet]
• [MobileFaceNet]

ShuffleNet v2

• ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design
• [ShuffleNet v2]
• [ShuffleNet v2]

MnasNet

• MnasNet: Platform-Aware Neural Architecture Search for Mobile
• [MnasNet]

MobileNet v3

• MobileNet v3 : Searching for MobileNetV3
• [MobileNet v3]

MixNet

• MixNet: Mixed Depthwise Convolutional Kernels
• [MixNet]