Improving Sketch Colorization using Adversarial Segmentation Consistency
This addresses the problem of generating color images from sketches for users in creative or design fields, offering an incremental improvement over existing methods by removing the need for paired data or additional user input.
The paper tackles sketch colorization by introducing an adversarial segmentation consistency loss that leverages semantic segmentation from a panoptic network, enabling unpaired translation without requiring segmentation labels. The method improves the baseline by up to 35 points on the FID metric across four datasets.
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.