Unsupervised Attention-guided Image-to-Image Translation

Authors:
Youssef Alami Mejjati University of Bath
Christian Richardt University of Bath
James Tompkin Brown University
Darren Cosker University of Bath
Kwang In Kim University of Bath

Introduction:

Current unsupervised image-to-image translation techniques struggle to focus their attention on individual objects without altering the background or the way multiple objects interact within a scene.

Abstract:

Current unsupervised image-to-image translation techniques struggle to focus their attention on individual objects without altering the background or the way multiple objects interact within a scene. Motivated by the important role of attention in human perception, we tackle this limitation by introducing unsupervised attention mechanisms which are jointly adversarially trained with the generators and discriminators. We empirically demonstrate that our approach is able to attend to relevant regions in the image without requiring any additional supervision, and that by doing so it achieves more realistic mappings compared to recent approaches.

You may want to know: