MD-GAN: Multi-Scale Diversity GAN for Large Masks Inpainting
Open Access
- 29 May 2025
- journal article
- research article
- Published by MDPI AG in Electronics
- Vol. 14 (11) , 2218
Abstract
Image inpainting approaches have made considerable progress with the assistance of generative adversarial networks (GANs) recently. However, current inpainting methods are incompetent in handling the cases with large masks and they generally suffer from unreasonable structure. We find that the main reason is the lack of an effective receptive field in the inpainting network. To alleviate this issue, we propose a new two-stage inpainting model called MD-GAN, which is a multi-scale diverse GAN. We inject dense combinations of dilated convolutions in multiple scales of inpainting networks to obtain more effective receptive fields. In fact, the result of inpainting large masks is generally not uniquely deterministic. To this end, we newly propose the multi-scale probabilistic diverse module, which achieves diverse content generation by spatial-adaptive normalization. Meanwhile, the convolutional block attention module is introduced to improve the ability to extract complex features. Perceptual diversity loss is added to enhance diversity. Extensive experiments on benchmark datasets including CelebA-HQ, Places2 and Paris Street View demonstrate that our approach is able to effectively inpaint diverse and structurally reasonable images.Keywords
Funding Information
- Key research Project of higher education institutions in Henan Province (24A520018, 62072160)
- National Natural Science Foundation of China (24A520018, 62072160)
This publication has 21 references indexed in Scilit:
- PD-GAN: Probabilistic Diverse GAN for Image Inpainting2021
- A comprehensive review of past and present image inpainting methodsComputer Vision and Image Understanding, 2021
- Pluralistic Image Completion2019
- Learning Pyramid-Context Encoder Network for High-Quality Image Inpainting2019
- Semantic Image Synthesis With Spatially-Adaptive Normalization2019
- Occluded offline handwritten Chinese character recognition using deep convolutional generative adversarial network and improved GoogLeNetNeural Computing and Applications, 2018
- Image Inpainting for Irregular Holes Using Partial Convolutions2018
- CBAM: Convolutional Block Attention Module2018
- Context Encoders: Feature Learning by Inpainting2016
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015