Image Model Blocks

Edit

Computer Vision • 93 methods

Image Model Blocks are building blocks used in image models such as convolutional neural networks. Below you can find a continuously updating list of image model blocks.

Methods

Add a Method

Method	Year	Papers
Residual Block Deep Residual Learning for Image Recognition	2015	2534
Bottleneck Residual Block Deep Residual Learning for Image Recognition	2015	1934
Dense Block Densely Connected Convolutional Networks	2016	405
Squeeze-and-Excitation Block Squeeze-and-Excitation Networks	2017	388
Inception Module Going Deeper with Convolutions	2014	173
Spatial Attention Module CBAM: Convolutional Block Attention Module	2018	160
Non-Local Block Non-local Neural Networks	2017	158
Spatial Transformer Spatial Transformer Networks	2015	141
ResNeXt Block Aggregated Residual Transformations for Deep Neural Networks	2016	123
Fire Module SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size	2016	83
Inception-v3 Module Rethinking the Inception Architecture for Computer Vision	2015	82
PnP PnP-DETR: Towards Efficient Visual Analysis with Transformers	2021	59
Channel Attention Module CBAM: Convolutional Block Attention Module	2018	55
Wide Residual Block Wide Residual Networks	2016	52
Axial Attention Axial Attention in Multidimensional Transformers	2019	49
ShuffleNet Block ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices	2017	45
ESP ESPNet: Efficient Spatial Pyramid of Dilated Convolutions for Semantic Segmentation	2018	32
Hourglass Module Stacked Hourglass Networks for Human Pose Estimation	2016	30
SRGAN Residual Block Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network	2016	27
Reduction-A Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning	2016	22
Ghost Module GhostNet: More Features from Cheap Operations	2019	22
ENet Initial Block ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation	2016	21
ENet Bottleneck ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation	2016	21
ENet Dilated Bottleneck ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation	2016	21
Res2Net Block Res2Net: A New Multi-scale Backbone Architecture	2019	19
Ghost Bottleneck GhostNet: More Features from Cheap Operations	2019	18
ShuffleNet V2 Block ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design	2018	18
Split Attention ResNeSt: Split-Attention Networks	2020	15
Selective Kernel Selective Kernel Networks	2019	15
DPN Block Dual Path Networks	2017	15
Inception-ResNet-v2-C Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning	2016	14
Inception-ResNet-v2-B Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning	2016	14
Inception-ResNet-v2-A Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning	2016	14
Inception-ResNet-v2 Reduction-B Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning	2016	14
CBAM CBAM: Convolutional Block Attention Module	2018	14
NEAT NEAT: Neural Attention Fields for End-to-End Autonomous Driving	2021	13
Efficient Channel Attention ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks	2019	12
ShuffleNet V2 Downsampling Block ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design	2018	12
FBNet Block FBNet: Hardware-Aware Efficient ConvNet Design via Differentiable Neural Architecture Search	2018	12
Inception-A Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning	2016	11
Inception-B Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning	2016	11
Inception-C Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning	2016	11
Reduction-B Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning	2016	11
Global Context Block GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond	2019	11
One-Shot Aggregation An Energy and GPU-Computation Efficient Backbone Network for Real-Time Object Detection	2019	9
Pyramidal Residual Unit Deep Pyramidal Residual Networks	2016	8
Pyramidal Bottleneck Residual Unit	2000	8
Fractal Block FractalNet: Ultra-Deep Neural Networks without Residuals	2016	6
TNT Transformer in Transformer	2021	6
Dilated Bottleneck Block DetNet: A Backbone network for Object Detection	2018	5
Dilated Bottleneck with Projection Block DetNet: A Backbone network for Object Detection	2018	5
NVAE Generative Residual Cell NVAE: A Deep Hierarchical Variational Autoencoder	2020	5
NVAE Encoder Residual Cell NVAE: A Deep Hierarchical Variational Autoencoder	2020	5
Bottleneck Transformer Block Bottleneck Transformers for Visual Recognition	2021	5
Spatial Feature Transform Recovering Realistic Texture in Image Super-resolution by Deep Spatial Feature Transform	2018	5
Big-Little Module Big-Little Net: An Efficient Multi-Scale Feature Representation for Visual and Speech Recognition	2018	4
Scale Aggregation Block Data-Driven Neuron Allocation for Scale Aggregation Networks	2019	4
Multiscale Dilated Convolution Block Neural Photo Editing with Introspective Adversarial Networks	2016	4
XCiT Layer XCiT: Cross-Covariance Image Transformers	2021	4
Local Patch Interaction XCiT: Cross-Covariance Image Transformers	2021	4
Mixer Layer MLP-Mixer: An all-MLP Architecture for Vision	2021	4
Style-based Recalibration Module SRM : A Style-based Recalibration Module for Convolutional Neural Networks	2019	3
Two-Way Dense Layer Pelee: A Real-Time Object Detection System on Mobile Devices	2018	3
SqueezeNeXt Block SqueezeNext: Hardware-Aware Neural Network Design	2018	3
EESP ESPNetv2: A Light-weight, Power Efficient, and General Purpose Convolutional Neural Network	2018	2
CSPResNeXt Block CSPNet: A New Backbone that can Enhance Learning Capability of CNN	2019	2
Elastic Dense Block ELASTIC: Improving CNNs with Dynamic Scaling Policies	2018	2
DVD-GAN GBlock Adversarial Video Generation on Complex Datasets	2019	2
DVD-GAN DBlock Adversarial Video Generation on Complex Datasets	2019	2
LRNet Local Relation Networks for Image Recognition	2019	2
Harmonic Block Harmonic Convolutional Networks based on Discrete Cosine Transform	2020	1
Spatial Group-wise Enhance Spatial Group-wise Enhance: Improving Semantic Feature Learning in Convolutional Networks	2019	1
Residual SRM SRM : A Style-based Recalibration Module for Convolutional Neural Networks	2019	1
DiCE Unit DiCENet: Dimension-wise Convolutions for Efficient Networks	2019	1
DimFuse DiCENet: Dimension-wise Convolutions for Efficient Networks	2019	1
Strided EESP ESPNetv2: A Light-weight, Power Efficient, and General Purpose Convolutional Neural Network	2018	1
Compact Global Descriptor Compact Global Descriptor for Neural Networks	2019	1
OSA (identity mapping + eSE) CenterMask : Real-Time Anchor-Free Instance Segmentation	2019	1
Effective Squeeze-and-Excitation Block CenterMask : Real-Time Anchor-Free Instance Segmentation	2019	1
Elastic ResNeXt Block ELASTIC: Improving CNNs with Dynamic Scaling Policies	2018	1
Depthwise Fire Module CornerNet-Lite: Efficient Keypoint Based Object Detection	2019	1
CornerNet-Squeeze Hourglass Module CornerNet-Lite: Efficient Keypoint Based Object Detection	2019	1
Mobile DenseNet EfficientPose: Scalable single-person pose estimation	2020	1
E-MBConv EfficientPose: Scalable single-person pose estimation	2020	1
Contextual Residual Aggregation Contextual Residual Aggregation for Ultra High-Resolution Image Inpainting	2020	1
Content-Conditioned Style Encoder COCO-FUNIT: Few-Shot Unsupervised Image Translation with a Content Conditioned Style Encoder	2020	1
Hierarchical-Split Block HS-ResNet: Hierarchical-Split Block on Convolutional Neural Network	2020	1
AttLWB Liquid Warping GAN with Attention: A Unified Framework for Human Image Synthesis	2020	1
Patch Merger Learning to Merge Tokens in Vision Transformers	2022	1
Global Local Attention Module All the attention you need: Global-local, spatial-channel attention for image retrieval	2021	1

Image Model Blocks Edit

Methods Add a Method

Image Model Blocks

Edit

Methods

Add a Method