TL;DR: A layered recursive GAN for image generation, which considers the structure in images and can disentangle the foreground objects from background well in unsupervised manner.
Abstract: We present LR-GAN: an adversarial image generation model which takes scene structure and context into account. Unlike previous generative adversarial networks (GANs), the proposed GAN learns to generate image background and foregrounds separately and recursively, and stitch the foregrounds on the background in a contextually relevant manner to produce a complete natural image. For each foreground, the model learns to generate its appearance, shape and pose. The whole model is unsupervised, and is trained in an end-to-end manner with conventional gradient descent methods. The experiments demonstrate that LR-GAN can generate more natural images with objects that are more human recognizable than baseline GANs.
Keywords: Computer vision, Deep learning, Unsupervised Learning
Conflicts: fb.com, vt.edu, gatech.edu