Improving Sketch Colorization using Adversarial Segmentation Consistency
Preprint
Abstract
We propose a new method for producing color images from sketches. Current solutions in sketch colorization either necessitate additional user instruction or are restricted to the "paired" translation strategy. We leverage semantic image segmentation from a general-purpose panoptic segmentation network to generate an additional adversarial loss function. The proposed loss function is compatible with any GAN model. Our method is not restricted to datasets with segmentation labels and can be applied to unpaired translation tasks as well. Using qualitative, and quantitative analysis, and based on a user study, we demonstrate the efficacy of our method on four distinct image datasets. On the FID metric, our model improves the baseline by up to 35 points. Our code, pretrained models, scripts to produce newly introduced datasets and corresponding sketch images are available at https://github.com/giddyyupp/AdvSegLoss.Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- Dual Color Space Guided Sketch ColorizationIEEE Transactions on Image Processing, 2021
- Unsupervised Sketch to Photo Synthesis2020
- GANILLA: Generative adversarial networks for image to illustration translationImage and Vision Computing, 2020
- Semantic Understanding of Scenes Through the ADE20K DatasetInternational Journal of Computer Vision, 2018
- Image Generation from Sketch Constraint Using Contextual GAN2018
- Auto-painter: Cartoon image generation from sketch by using conditional Wasserstein generative adversarial networksNeurocomputing, 2018
- DualGAN: Unsupervised Dual Learning for Image-to-Image Translation2017
- Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks2017
- Microsoft COCO: Common Objects in Context2014
- XDoG: An eXtended difference-of-Gaussians compendium including advanced image stylizationComputers & Graphics, 2012