Scared11a/FCCNs

FCCNs: Fully Complex-valued Convolutional Networks using Complex-valued Color Model and Loss Function

we propose a complex-valued color model and loss function and turn fully-connected layers into convolutional layers. All these contributions culminate in what we call FCCNs (Fully Complex-valued Convolutional Networks), which take complex-valued images as inputs, perform only complex-valued operations, and have a complex-valued loss function. Here is the link to the pre-trained models on ImageNet.

This repo contains code for the paper "FCCNs: Fully Complex-valued Convolutional Networks usnig Complex-valued Color Model and Loss Function".

@INPROCEEDINGS{10377516,
  author={Yadav, Saurabh and Jerripothula, Koteswar Rao},
  booktitle={2023 IEEE/CVF International Conference on Computer Vision (ICCV)}, 
  title={FCCNs: Fully Complex-valued Convolutional Networks using Complex-valued Color Model and Loss Function}, 
  year={2023},
  volume={},
  number={},
  pages={10655-10664},
  keywords={Convolutional codes;Computer vision;Image color analysis;Computational modeling;Benchmark testing;Convolutional neural networks;Image classification},
  doi={10.1109/ICCV51070.2023.00981}}