Semantic Image Inpainting with Deep Generative Models
This addresses the problem of generating realistic content for large holes in images, which is incremental as it builds on existing generative models but introduces a novel inference approach.
The paper tackles semantic image inpainting for large missing regions by proposing a method that uses a trained generative model to infer missing content based on context and prior losses, achieving pixel-level photorealism and significantly outperforming state-of-the-art methods on three datasets.
Semantic image inpainting is a challenging task where large missing regions have to be filled based on the available visual data. Existing methods which extract information from only a single image generally produce unsatisfactory results due to the lack of high level context. In this paper, we propose a novel method for semantic image inpainting, which generates the missing content by conditioning on the available data. Given a trained generative model, we search for the closest encoding of the corrupted image in the latent image manifold using our context and prior losses. This encoding is then passed through the generative model to infer the missing content. In our method, inference is possible irrespective of how the missing content is structured, while the state-of-the-art learning based method requires specific information about the holes in the training phase. Experiments on three datasets show that our method successfully predicts information in large missing regions and achieves pixel-level photorealism, significantly outperforming the state-of-the-art methods.