Interactive Sketch & Fill: Multiclass Sketch-to-Image Translation

A new GAN-based interactive image generation system for drawing objects from scratch that: 1) generates full images given partial user strokes (or sketches); 2) serves as a recommender system that suggests or helps the user during their creative process to help them generate a desired image; and 3) uses a single conditional GAN model for multiple image classes, via a gating-based conditioning mechanism.

一款基于生成对抗网络（Generative Adversarial Network，GAN）的交互式图像生成系统，可从零开始绘制物体，具备以下三项特性：1) 可根据用户提供的局部手绘笔触（或草图）生成完整图像；2) 可作为创作推荐系统，在用户的创作过程中提供建议与辅助，协助其生成符合预期的目标图像；3) 通过基于门控的调节机制，仅使用单个条件生成对抗网络（Conditional GAN）即可适配多类图像。