CRD-CGAN: Category-Consistent and Relativistic Constraints for Diverse Text-to-Image Generation
This work addresses the challenge of diverse text-to-image generation for computer vision applications, representing an incremental improvement over existing GAN-based methods.
The paper tackles the problem of generating diverse and photo-realistic images from text descriptions by introducing category-consistent and relativistic constraints, resulting in a method that outperforms state-of-the-art approaches on datasets like Birds-200-2011, Oxford-102 flower, and MSCOCO 2014.
Generating photo-realistic images from a text description is a challenging problem in computer vision. Previous works have shown promising performance to generate synthetic images conditional on text by Generative Adversarial Networks (GANs). In this paper, we focus on the category-consistent and relativistic diverse constraints to optimize the diversity of synthetic images. Based on those constraints, a category-consistent and relativistic diverse conditional GAN (CRD-CGAN) is proposed to synthesize $K$ photo-realistic images simultaneously. We use the attention loss and diversity loss to improve the sensitivity of the GAN to word attention and noises. Then, we employ the relativistic conditional loss to estimate the probability of relatively real or fake for synthetic images, which can improve the performance of basic conditional loss. Finally, we introduce a category-consistent loss to alleviate the over-category issues between K synthetic images. We evaluate our approach using the Birds-200-2011, Oxford-102 flower and MSCOCO 2014 datasets, and the extensive experiments demonstrate superiority of the proposed method in comparison with state-of-the-art methods in terms of photorealistic and diversity of the generated synthetic images.