Thanks for sharing this work. I am a student and interested in this job.
I have some questions about the details of the network.
(1)In Fig2 (b), why does an hk-1 pass back behind the RMU module?
(2)How does the dimension of hidden variable (hk) through RMU change?
(3)In Eq.(10), fl and fh are mentioned. A bilinear upsampling is chosen as fl. What does fh represent?
(4)would your like to share network details such as the strides of convolutions and the channels of inputs and outputs of each module in the paper.
Dear Tak-Wai Hui,
Thanks for sharing this work. I am a student and interested in this job. I have some questions about the details of the network.
(1)In Fig2 (b), why does an hk-1 pass back behind the RMU module? (2)How does the dimension of hidden variable (hk) through RMU change? (3)In Eq.(10), fl and fh are mentioned. A bilinear upsampling is chosen as fl. What does fh represent? (4)would your like to share network details such as the strides of convolutions and the channels of inputs and outputs of each module in the paper.
Looking forward for your reply.