Learning Foreground-Background Segmentation from Improved Layered GANs
This addresses the need for cheaper and more efficient training data in computer vision, particularly for image segmentation tasks, but it is incremental as it builds on existing layered GAN approaches.
The paper tackles the problem of expensive human supervision for image segmentation by proposing a method to automatically synthesize paired photo-realistic images and segmentation masks for training foreground-background segmentation networks. The result is competitive generation quality and segmentation performance on single-object datasets compared to related methods.
Deep learning approaches heavily rely on high-quality human supervision which is nonetheless expensive, time-consuming, and error-prone, especially for image segmentation task. In this paper, we propose a method to automatically synthesize paired photo-realistic images and segmentation masks for the use of training a foreground-background segmentation network. In particular, we learn a generative adversarial network that decomposes an image into foreground and background layers, and avoid trivial decompositions by maximizing mutual information between generated images and latent variables. The improved layered GANs can synthesize higher quality datasets from which segmentation networks of higher performance can be learned. Moreover, the segmentation networks are employed to stabilize the training of layered GANs in return, which are further alternately trained with Layered GANs. Experiments on a variety of single-object datasets show that our method achieves competitive generation quality and segmentation performance compared to related methods.