PyTorch implementation of 3D segmentation models
Summary (KOR): https://bo-10000.tistory.com/39?category=1054227
Applied residual learning. Consists of stacked residual modules(b) and used 4 auxiliary classifiers.
https://arxiv.org/pdf/1608.05895.pdf
Used Attention Gate, which suppress irrelevant regions in an input image while highlighting salient features useful for a specific task.
https://arxiv.org/pdf/1804.03999.pdf
Similar to U-Net structure. Applied residual connection at each stage and used PReLU for activation function.
https://arxiv.org/pdf/1606.04797.pdf
