Semantic Segmentation using Generative Adversarial Network
Wenxin Chen, Ting Zhang, Xing Zhao · 2021
In the field of deep learning, semantic segmentation is a classical computer vision problem. Generative adversarial network is composed of generator and discriminator, which shows excellent performance in various generation tasks. In order to improve the segmentation effect of the model further, a generative adversarial network for semantic segmentation is proposed in this paper. By introducing the idea of patch discriminant, the model can achieve a balance between the global discriminant ability and the detail discriminant ability. Experiments in CamVid and Cityscapes datasets show that this model can effectively improve the accuracy of semantic segmentation.