How is the idea any different?

[srinadhu]

Hello @gjy3035,

I skimmed through the paper. Please do correct me if I am wrong, How is the attention presented in here any different from Dual Attention Network for Scene Segmentation. They look same to me. Thanks

[gjy3035]

Firstly, comparing the attention modules, SCAR is almost the same as DA Net.

Then, we need to make explanations for it: Our original paper is inspired by SCA cnn (ICCV2017) and non-local (ICCV2017). It is completed in Sep. 2018 and initially submitted to ICASSP (2018.10) but rejected by a wired reason. The record is shown as follows:

About after CVPR's deadline last year, I noted this paper, and I find we crashed. By comparing the network, there are some differences as follows: backbone, the fusion of attention models and applications.

Finally, thanks for your attention!