Semi-supervised Image Semantic Segmentation Based on Adaptive Adversarial Learning

The fixed penalty factor is introduced to the model in the existed adaptive adversarial learning of different feature layers, and the FCN (Fully Convolutional Networks) is used as the basis framework of discriminators, which will result in low generalization ability of the model and the segmentation result is not as fine as expected. Furthermore, it is easy to cause the problem of class infection and class drift, which exists in the latest domain adaptive method in more complex scenarios. To address this issue, a new semi-supervised semantics segmentation method based on GAN network is proposed for urban scene segmentation in this paper. We use architecture similar to SegNet and utilize all fully convolutional layers to retain the spatial information, which adopts maximum pooling method to up-sample image nonlinearly, which carries on the advantage of FCN to process any scale input image as well as to preserve refined feature correlation information. An adaptive learning rate is employed to adjust deep neural network features in different layers in the processing of fusion adversarial loss and cross-entropy loss. So the presented GAN model improves semantic segmentation precision, which is capable of adjusting the weight of adversarial loss and cross-entropy loss through adaptive learning rate to update the segmentation net in the generator. Moreover, the segmentation result refined by the provided model, because SegNet takes the place of FCN in the discriminator part of GAN to get rid of violence pooling, edge information of unlabeled target dataset is imported in the networks. As a result, the margin area in the net is corrected effectively and the edge information in the image is preserved as far as possible. The validation results on PASCAL VOC2012 standard data set show that the proposed model is capable of segmenting object in more complex scenarios, relieving class infections and class drift, moreover enhance the edge detail effectively.