This work proposes a effective CNN based on 3DUnet, CAAL-Unet. It uses 3D partical Convolution to replace the traditional convolution, thereby reducing network parameters and redundant feature maps.

Abstract: Hyperspectral image denoising is crucial for accurate extraction of spectral information. However, current convolutional neural network (CNN)-based methods have inherent limitations, while ...

This repo is the source code for BiTr-Unet: a CNN-Transformer Combined Network for MRI Brain Tumor Segmentation. The dataset that this model is designed for is BraTS ...

def build_unet_1d(n_times, n_feature, n_out, units, drop=0.5): Builds a 1D U-Net model for time series data with batch normalization and sigmoid activation for the output layer. Parameters: - n_times: ...

a streamlined novel precipitation nowcasting method built on the UNet architecture. RainHCNet incorporates a hybrid channel-spatial attention mechanism to effectively capture low-frequency information ...