文件名称：Unsupervised-Attention-guided-Image-to-Image-Translation-master.zip
文件大小：747KB
文件格式：ZIP
更新时间：2023-01-03 18:33:20
attention-guided
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.
【文件预览】：
Unsupervised-Attention-guided-Image-to-Image-Translation-master
----download_datasets.sh(814B)
----main.py(24KB)
----cyclegan_datasets.py(1KB)
----Trained_models()
--------README.md(311B)
----test()
--------evaluate_networks.py(1KB)
--------__init__.py(0B)
--------test_model.py(1KB)
--------evaluate_losses.py(1KB)
--------test_losses.py(2KB)
----data_loader.py(3KB)
----LICENSE(1KB)
----imgs()
--------attentionMaps.jpg(257KB)
--------AtO.jpg(112KB)
--------ZtH.jpg(96KB)
--------OtA.jpg(81KB)
--------HtZ.jpg(76KB)
--------AGGANDiagram.jpg(155KB)
----losses.py(3KB)
----__init__.py(0B)
----create_cyclegan_dataset.py(2KB)
----configs()
--------exp_04.json(252B)
--------exp_04_test.json(140B)
--------exp_02.json(254B)
--------exp_01.json(252B)
--------exp_05.json(264B)
--------exp_02_test.json(141B)
--------exp_05_test.json(151B)
--------exp_01_test.json(141B)
----model.py(11KB)
----layers.py(6KB)
----README.md(4KB)