reviseFor the RBA-Net the backbone of encoder network is ResNet-50 pre-trained on ImageNet The batch size is set to 8 and Adam optimizer is adopted for training network according to the end-to-end way

The RBA-Net utilizes ResNet-50 as the backbone of its encoder network, which has been pre-trained on ImageNet. During training, a batch size of 8 is used, and the network is optimized using the Adam optimizer in an end-to-end manner. The initial learning rate is set to 0.0001, and a "poly" learning rate policy is employed for all experiments. The learning rate is multiplied by MM after each iteration and training is terminated at 100 epochs.