- [IJCAI 2019] Learning to Draw Text in Natural Images with Conditional Adversarial Networks
STS-GAN，生成文本图像。
Inthiswork,weproposeanentirelylearning-based method to automatically synthesize text sequence in natural images leveraging conditional adversarial networks. As vanilla GANs are clumsy to capture structural text patterns, directly employing GANs for text image synthesis typically results in illegible images. Therefore, we design a two-stage architecture to generate repeated characters in images. Firstly, a character generator attempts to synthesize local character appearance independently, so that the legible characters in sequence can be obtained. To achieve style consistency of characters, we propose a novel style lossbasedonvariance-minimization. Secondly,we design a pixel-manipulation word generator constrainedbyself-regularization,whichlearnstoconvert local characters to plausible word image. Experiments on SVHN dataset and ICDAR, IIIT5K datasets demonstrate our method is able to synthesize visually appealing text images. Besides, we also show the high-quality images synthesized by our method can be used to boost the performance of a scene text recognition algorithm.
在这项工作中，我们提出了一种基于整体学习的方法，利用条件对抗性网络自动合成自然图像中的文本序列。由于普通的甘值无法捕捉结构化文本模式，直接使用甘值进行文本图像合成通常会导致图像难以辨认。因此，我们设计了一个两阶段的架构来生成图像中的重复字符。首先，字符生成器尝试独立地综合局部字符的外观，从而获得序列清晰的字符。为了实现字符的风格一致性，我们提出了一种基于方差最小化的新风格。其次，设计了一种基于自正则化约束的像素操作字生成器，学习将局部字符转换为可信的字图像。在SVHN数据集和ICDAR、IIIT5K数据集上的实验表明，该方法能够综合出具有视觉吸引力的文本图像。此外，我们还证明了用我们的方法合成的高质量图像可以提高场景文本识别算法的性能。