Multimodal Image Outpainting With Regularized Normalized Diversification
This addresses the challenge of mode collapse in conditional image synthesis for applications like content creation, though it is incremental as it builds on existing methods.
The paper tackled the problem of generating diverse and realistic backgrounds from a small foreground region, known as image outpainting, by proposing a regularization method and feature pyramid discriminator, resulting in more diverse images without sacrificing visual quality compared to state-of-the-art approaches on datasets like CelebA and Cityscape.
In this paper, we study the problem of generating a set ofrealistic and diverse backgrounds when given only a smallforeground region. We refer to this task as image outpaint-ing. The technical challenge of this task is to synthesize notonly plausible but also diverse image outputs. Traditionalgenerative adversarial networks suffer from mode collapse.While recent approaches propose to maximize orpreserve the pairwise distance between generated sampleswith respect to their latent distance, they do not explicitlyprevent the diverse samples of different conditional inputsfrom collapsing. Therefore, we propose a new regulariza-tion method to encourage diverse sampling in conditionalsynthesis. In addition, we propose a feature pyramid dis-criminator to improve the image quality. Our experimen-tal results show that our model can produce more diverseimages without sacrificing visual quality compared to state-of-the-arts approaches in both the CelebA face dataset and the Cityscape scene dataset.