Durall, RicardRicardDurallPfreundt, Franz-JosefFranz-JosefPfreundtKöthe, UllrichUllrichKötheKeuper, JanisJanisKeuper2022-03-142022-03-142019https://publica.fraunhofer.de/handle/publica/40834410.1007/978-3-030-33676-9_21Recent deep learning based approaches have shown remarkable success on object segmentation tasks. However, there is still room for further improvement. Inspired by generative adversarial networks, we present a generic end-to-end adversarial approach, which can be combined with a wide range of existing semantic segmentation networks to improve their segmentation performance. The key element of our method is to replace the commonly used binary adversarial loss with a high resolution pixel-wise loss. In addition, we train our generator employing stochastic weight averaging fashion, which further enhances the predicted output label maps leading to state-of-the-art results. We show, that this combination of pixel-wise adversarial training and weight averaging leads to significant and consistent gains in segmentation performance, compared to the baseline models.en003006519conference paper