In these days, the "feature fusion" in deep neural networks became so popular that many of the academic papers implement it without references.
So, could you please recommend some papers dealing with the effects of feature fusion (i.e., concatenating the hidden layers/multiple inputs), or forking the network (for multiple outputs) ?
Papers in any fields are welcome.
Appreciated for your attention.
Have a look at https://cse.hkust.edu.hk/faculty/jamesk/papers/prl02.pdf
https://openaccess.thecvf.com/content_ECCV_2018/papers/Xiaoqing_Ye_3D_Recurrent_Neural_ECCV_2018_paper.pdf
Article Feature Fusion Based on Convolutional Neural Network for SAR ATR
this paper is published here in Researchgate
"Feature Fusion Based on Convolutional Neural Network for SAR ATR"
I hope these articles help you
https://www.hindawi.com/journals/je/2020/7286187/
Article Multi-Layers Feature Fusion of Convolutional Neural Network ...
Conference Paper Hierarchical Features Fusion for Image Aesthetics Assessment
Article Blind Image Quality Assessment Using A Deep Bilinear Convolu...
Preprint Blindly Assess Quality of In-the-Wild Videos via Quality-awa...
Article Dual Head Network for No-Reference Quality Assessment Toward...
I have a paper on Feature Fusion directly in the deep network in the DWT domain. Without more information on exactly what you need, I cannot recommend any more specific works. Hope this helps
Preprint A Novel Deep Learning Method for Thermal to Annotated Therma...
- Badges
- Science topic