# PD-GAN: Probabilistic Diverse GAN for Image Inpainting

arXiv: https://arxiv.org/abs/2105.02201 · HTML: https://mlsanity.com/paper/2105.02201/

## Key facts

- Cited by 80 papers in the map (data through 14 Oct 2026).
- Builds on 44 of its 60 references that are in the map.
- First posted on arXiv 5 May 2021.

## Builds on (most cited first, 30 of 44)

- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [Perceptual Losses for Real-Time Style Transfer and Super-Resolution](https://mlsanity.com/paper/1603.08155/) (2016)
- [Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization](https://mlsanity.com/paper/1703.06868/) (2017)
- [Group Normalization](https://mlsanity.com/paper/1803.08494/) (2018)
- [Context Encoders: Feature Learning by Inpainting](https://mlsanity.com/paper/1604.07379/) (2016)
- [Semantic Image Synthesis with Spatially-Adaptive Normalization](https://mlsanity.com/paper/1903.07291/) (2019)
- [Instance Normalization: The Missing Ingredient for Fast Stylization](https://mlsanity.com/paper/1607.08022/) (2016)
- [High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs](https://mlsanity.com/paper/1711.11585/) (2017)
- [Multi-Scale Context Aggregation by Dilated Convolutions](https://mlsanity.com/paper/1511.07122/) (2015)
- [Multimodal Unsupervised Image-to-Image Translation](https://mlsanity.com/paper/1804.04732/) (2018)
- [MaskGAN: Towards Diverse and Interactive Facial Image Manipulation](https://mlsanity.com/paper/1907.11922/) (2019)
- [Free-Form Image Inpainting with Gated Convolution](https://mlsanity.com/paper/1806.03589/) (2018)
- [Generative Image Inpainting with Contextual Attention](https://mlsanity.com/paper/1801.07892/) (2018)
- [Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks](https://mlsanity.com/paper/1602.07868/) (2016)
- [Image Inpainting for Irregular Holes Using Partial Convolutions](https://mlsanity.com/paper/1804.07723/) (2018)
- [Toward Multimodal Image-to-Image Translation](https://mlsanity.com/paper/1711.11586/) (2017)
- [VITON: An Image-based Virtual Try-on Network](https://mlsanity.com/paper/1711.08447/) (2017)
- [Geometric GAN](https://mlsanity.com/paper/1705.02894/) (2017)
- [EdgeConnect: Generative Image Inpainting with Adversarial Edge Learning](https://mlsanity.com/paper/1901.00212/) (2019)
- [Pluralistic Image Completion](https://mlsanity.com/paper/1903.04227/) (2019)
- [Mode Seeking Generative Adversarial Networks for Diverse Image Synthesis](https://mlsanity.com/paper/1903.05628/) (2019)
- [Contextual Residual Aggregation for Ultra High-Resolution Image Inpainting](https://mlsanity.com/paper/2005.09704/) (2020)
- [StructureFlow: Image Inpainting via Structure-aware Appearance Flow](https://mlsanity.com/paper/1908.03852/) (2019)
- [Cross-domain Correspondence Learning for Exemplar-based Image Translation](https://mlsanity.com/paper/2004.05571/) (2020)
- [Recurrent Feature Reasoning for Image Inpainting](https://mlsanity.com/paper/2008.03737/) (2020)
- [Rethinking Image Inpainting via a Mutual Encoder-Decoder with Feature Equalizations](https://mlsanity.com/paper/2007.06929/) (2020)
- [Coherent Semantic Attention for Image Inpainting](https://mlsanity.com/paper/1905.12384/) (2019)
- [Towards Multi-pose Guided Virtual Try-on Network](https://mlsanity.com/paper/1902.11026/) (2019)

## Built on by (most cited first, 30 of 80)

- [BrushNet: A Plug-and-Play Image Inpainting Model with Decomposed Dual-Branch Diffusion](https://mlsanity.com/paper/2403.06976/) (2024)
- [Depth-Aware Generative Adversarial Network for Talking Head Video Generation](https://mlsanity.com/paper/2203.06605/) (2022)
- [Is Sora a World Simulator? A Comprehensive Survey on General World Models and Beyond](https://mlsanity.com/paper/2405.03520/) (2024)
- [A Complete Survey on Generative AI (AIGC): Is ChatGPT from GPT-4 to GPT-5 All You Need?](https://mlsanity.com/paper/2303.11717/) (2023)
- [InfiniteNature-Zero: Learning Perpetual View Generation of Natural Scenes from Single Images](https://mlsanity.com/paper/2207.11148/) (2022)
- [Follow-Your-Creation: Empowering 4D Creation through Video Inpainting](https://mlsanity.com/paper/2506.04590/) (2025)
- [Persistent Nature: A Generative Model of Unbounded 3D Worlds](https://mlsanity.com/paper/2303.13515/) (2023)
- [Deep Learning-based Image and Video Inpainting: A Survey](https://mlsanity.com/paper/2401.03395/) (2024)
- [SketchEdit: Mask-Free Local Image Manipulation with Partial Sketches](https://mlsanity.com/paper/2111.15078/) (2021)
- [Bringing Old Films Back to Life](https://mlsanity.com/paper/2203.17276/) (2022)
- [BrushEdit: All-In-One Image Inpainting and Editing](https://mlsanity.com/paper/2412.10316/) (2024)
- [TGFuse: An Infrared and Visible Image Fusion Approach Based on Transformer and Generative Adversarial Network](https://mlsanity.com/paper/2201.10147/) (2022)
- [BIPS: Bi-modal Indoor Panorama Synthesis via Residual Depth-aided Adversarial Learning](https://mlsanity.com/paper/2112.06179/) (2021)
- [One Model to Edit Them All: Free-Form Text-Driven Image Manipulation with Semantic Modulations](https://mlsanity.com/paper/2210.07883/) (2022)
- [CharacterFactory: Sampling Consistent Characters with GANs for Diffusion Models](https://mlsanity.com/paper/2404.15677/) (2024)
- [RORem: Training a Robust Object Remover with Human-in-the-Loop](https://mlsanity.com/paper/2501.00740/) (2025)
- [RefFusion: Reference Adapted Diffusion Models for 3D Scene Inpainting](https://mlsanity.com/paper/2404.10765/) (2024)
- [Delving StyleGAN Inversion for Image Editing: A Foundation Latent Space Viewpoint](https://mlsanity.com/paper/2211.11448/) (2022)
- [HiScene: Creating Hierarchical 3D Scenes with Isometric View Generation](https://mlsanity.com/paper/2504.13072/) (2025)
- [FreGAN: Exploiting Frequency Components for Training GANs under Limited Data](https://mlsanity.com/paper/2210.05461/) (2022)
- [Human MotionFormer: Transferring Human Motions with Vision Transformers](https://mlsanity.com/paper/2302.11306/) (2023)
- [HS-Diffusion: Semantic-Mixing Diffusion for Head Swapping](https://mlsanity.com/paper/2212.06458/) (2022)
- [From Posterior Sampling to Meaningful Diversity in Image Restoration](https://mlsanity.com/paper/2310.16047/) (2023)
- [Improving Text-guided Object Inpainting with Semantic Pre-inpainting](https://mlsanity.com/paper/2409.08260/) (2024)
- [Generative Adversarial Networks for Image Super-Resolution: A Survey](https://mlsanity.com/paper/2204.13620/) (2022)
- [SIEDOB: Semantic Image Editing by Disentangling Object and Background](https://mlsanity.com/paper/2303.13062/) (2023)
- [Flux Already Knows -- Activating Subject-Driven Image Generation without Training](https://mlsanity.com/paper/2504.11478/) (2025)
- [Reference-Guided Large-Scale Face Inpainting with Identity and Texture Control](https://mlsanity.com/paper/2303.07014/) (2023)
- [Enhancement of Novel View Synthesis Using Omnidirectional Image Completion](https://mlsanity.com/paper/2203.09957/) (2022)
- [RDFC-GAN: RGB-Depth Fusion CycleGAN for Indoor Depth Completion](https://mlsanity.com/paper/2306.03584/) (2023)

## Abstract (arXiv)

We propose PD-GAN, a probabilistic diverse GAN for image inpainting. Given an input image with arbitrary hole regions, PD-GAN produces multiple inpainting results with diverse and visually realistic content. Our PD-GAN is built upon a vanilla GAN which generates images based on random noise. During image generation, we modulate deep features of input random noise from coarse-to-fine by injecting an initially restored image and the hole regions in multiple scales. We argue that during hole filling, the pixels near the hole boundary should be more deterministic (i.e., with higher probability trusting the context and initially restored image to create natural inpainting boundary), while those pixels lie in the center of the hole should enjoy more degrees of freedom (i.e., more likely to depend on the random noise for enhancing diversity). To this end, we propose spatially probabilistic diversity normalization (SPDNorm) inside the modulation to model the probability of generating a pixel conditioned on the context information. SPDNorm dynamically balances the realism and diversity inside the hole region, making the generated content more diverse towards the hole center and resemble neighboring image content more towards the hole boundary. Meanwhile, we propose a perceptual diversity loss to further empower PD-GAN for diverse content generation. Experiments on benchmark datasets including CelebA-HQ, Places2 and Paris Street View indicate that PD-GAN is effective for diverse and visually realistic image restoration.