Probabilistic Semantic Inpainting with Pixel Constrained CNNs

2018/10/08 by Emilien Dupont, Dupont, Emilien, Suhas Suresha +1 · 1 citation
Computer Science · #Advanced Image Processing Techniques #Computer Vision and Pattern Recognition (cs.CV) #Digital Media Forensic Detection #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · pdf · doi:10.48550/arxiv.1810.03728

openalex publication_date 2018/10/08 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Semantic inpainting is the task of inferring missing pixels in an image given surrounding pixels and high level image semantics. Most semantic inpainting algorithms are deterministic: given an image with missing regions, a single inpainted image is generated. However, there are often several plausible inpaintings for a given missing region. In this paper, we propose a method to perform probabilistic semantic inpainting by building a model, based on PixelCNNs, that learns a distribution of images conditioned on a subset of visible pixels. Experiments on the MNIST and CelebA datasets show that our method produces diverse and realistic inpaintings.

Citations

• The Cramer Distance as a Solution to Biased Wasserstein Gradients
• Globally and locally consistent image completion
• Generative Image Inpainting with Contextual Attention
• Generating Realistic Geology Conditioned on Physical Measurements with Generative Adversarial Networks
• Controllable Semantic Image Inpainting
• Conditional Image Generation with PixelCNN Decoders

Cited by

• DeepFlow: History Matching in the Space of Deep Generative Models

Related

• Semantic Image Inpainting with Deep Generative Models
• Fully Context-Aware Image Inpainting with a Learned Semantic Pyramid
• Image Inpainting via Iteratively Decoupled Probabilistic Modeling
• Controllable Semantic Image Inpainting
• Light-weight pixel context encoders for image inpainting