Painting Outside the Box: Image Outpainting with GANs

Abstract

The challenging task of image outpainting (extrapolation) has received comparatively little attention in relation to its cousin, image inpainting (completion). Accordingly, we present a deep learning approach based on Iizuka et al. for adversarially training a network to hallucinate past image boundaries. We use a three-phase training schedule to stably train a DCGAN architecture on a subset of the Places365 dataset. In line with Iizuka et al., we also use local discriminators to enhance the quality of our output. Once trained, our model is able to outpaint $128\times 128$ color images relatively realistically, thus allowing for recursive outpainting. Our results show that deep learning approaches to image outpainting are both feasible and promising.

arXiv comment5 pages, 9 figures

Cited by in corpus (10)

• Large Scale Image Completion via Co-Modulated Generative Adversarial Networks
• Sketch-Guided Scenery Image Outpainting
• Image Outpainting and Harmonization using Generative Adversarial Networks
• InfinityGAN: Towards Infinite-Pixel Image Synthesis
• COCO-GAN: Generation by Parts via Conditional Coordinating
• 180-degree Outpainting from a Single Image
• Enhanced Residual Networks for Context-based Image Outpainting
• In-N-Out: Towards Good Initialization for Inpainting and Outpainting
• Painting Outside as Inside: Edge Guided Image Outpainting via Bidirectional Rearrangement with Progressive Step Learning
• A self-adapting super-resolution structures framework for automatic design of GAN