CFNet
:: DESCRIPTION
CFNet combines a novel rotation equivariant convolution scheme, called conic convolution, and the DFT to aid networks in learning rotation-invariant tasks. This network has been especially designed to improve performance of CNNs on automated computational tasks related to microscopy image analysis.
::DEVELOPER
:: SCREENSHOTS
N/A
:: REQUIREMENTS
- Linux / Windows / MacOsX
- Python
:: DOWNLOAD
:: MORE INFORMATION
Citation
Bioinformatics. 2019 Jul 15;35(14):i530-i537. doi: 10.1093/bioinformatics/btz353.
Rotation equivariant and invariant neural networks for microscopy image analysis.
Chidester B, Zhou T, Do MN, Ma J.