Mustapha Oujaoura

XCC-Net: An X-Shaped Collective Convolution Network Architecture for Medical Image Segmentation

Anass Garbaz, Yassine Oukdach, Said Charfi, Mohamed El Ansari, Lahcen Koutti, Mustapha Hedabou, et al.

Encoder–decoder models are widely used for pixel-level segmentation due to their ability to capture and combine multiscale features. However, skip connections between the encoder and decoder often require cropping to mitigate border pixel loss during convolutions, which can intro…Read more