U Net
U-Net is a deep learning model used for image segmentation tasks, such as biomedical image segmentation. The structure of U-Net consists of two main parts: the encoder and the decoder.
Encoder: The encoder consists of a series of convolutional and max-pooling layers that reduce the spatial dimensions of the input image while increasing the number of feature maps. This process is repeated multiple times, resulting in a feature representation that summarizes the spatial information in the input image.
Decoder: The decoder takes the feature representation generated by the encoder and upscales it back to the original image resolution while increasing the number of feature maps. This is done using a series of transposed convolutions, which are essentially deconvolutional layers that increase the spatial resolution of the feature maps. Additionally, the decoder also uses skip connections that concatenate the corresponding feature maps from the encoder to the decoder, preserving low-level spatial information from the input image.
The U-Net architecture is symmetrical, with the encoder and decoder mirroring each other in terms of the number of layers and the type of layers used. The final layer of the decoder outputs a segmentation map, where each pixel is assigned a class label based on the activations of the corresponding feature map.
*Description created by ChatGPT
