特征融合热门方法及模型

特征融合方法（由深度之眼整理）

Concat（拼接）

1.Densenet

论文：Densely Connected Convolutional Networks

https://arxiv.org/abs/1608.06993

https://github.com/liuzhuang13/DenseNet

Add并行策略

2.Resnet

论文：Deep Residual Learning for Image Recognition

https://www.cv-foundation.org/openaccess/content_cvpr_2016/papers/He_Deep_Residual_Learning_CVPR_2016_paper.pdf

双线性池化

3.CBP

论文：Compact Bilinear Pooling

https://arxiv.org/abs/1511.06062

特征融合模型

1.SENet

论文：Squeeze-and-Excitation Networks

https://arxiv.org/abs/1709.01507

https://github.com/hujie-frank/SENet

2.GCNet

论文：Global Context Network

https://arxiv.org/abs/2012.13375

https://github.com/xvjiarui/GCNet

3.DANet

论文：Dual Attention Network for Scene Segmentation

https://arxiv.org/abs/1809.02983

https://github.com/junfu1115/DANet

4.PANet

论文：Path Aggregation Network for Instance Segmentation

https://arxiv.org/abs/1803.01534

https://github.com/ShuLiu1993/PANet

5.FPN

论文：Feature Pyramid Networks for Object Detection

https://arxiv.org/abs/1612.03144

6.ASPP

论文：DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs

https://arxiv.org/pdf/1606.00915.pdf

7.SPP-net

论文：Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition

https://arxiv.org/abs/1406.4729

8.PSP-net

论文：Pyramid Scene Parsing Network

https://arxiv.org/abs/1612.01105

9.ECA-Net

论文：ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks

https://arxiv.org/pdf/1910.03151.pdf

10.EffectiveSE

论文：CenterMask : Real-Time Anchor-Free Instance Segmentation

https://arxiv.org/pdf/1911.06667.pdf

https://github.com/youngwanLEE/CenterMask