SCSA-Net: Presentation of Two-View Reliable Correspondence Learning via Spatial-Channel Self-Attention
Pytorch implementation of SCSA-Net
Please use Python 3.6, opencv-contrib-python (3.4.0.12) and Pytorch (>= 1.1.0). Other dependencies should be easily installed through pip or conda.
If you need YFCC100M and SUN3D datasets, You can visit the code at https://github.com/zjhthu/OANet.git. We have uploaded the main code on 'core' folder.