Leveraging Visual Question Answering to Improve Text-to-Image Synthesis

Abstract

Generating images from textual descriptions has recently attracted a lot of interest. While current models can generate photo-realistic images of individual objects such as birds and human faces, synthesising images with multiple objects is still very difficult. In this paper, we propose an effective way to combine Text-to-Image (T2I) synthesis with Visual Question Answering (VQA) to improve the image quality and image-text alignment of generated images by leveraging the VQA 2.0 dataset. We create additional training samples by concatenating question and answer (QA) pairs and employ a standard VQA model to provide the T2I model with an auxiliary learning signal. We encourage images generated from QA pairs to look realistic and additionally minimize an external VQA loss. Our method lowers the FID from 27.84 to 25.38 and increases the R-prec. from 83.82% to 84.79% when compared to the baseline, which indicates that T2I synthesis can successfully be improved using a standard VQA model.

arXiv commentAccepted to the LANTERN workshop at COLING 2020

References in corpus (9)

• Conditional Generative Adversarial Nets
• GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
• Generative Adversarial Text to Image Synthesis
• Conditional Image Synthesis With Auxiliary Classifier GANs
• VQA: Visual Question Answering
• LXMERT: Learning Cross-Modality Encoder Representations from Transformers
• Semantic Object Accuracy for Generative Text-to-Image Synthesis
• Show, Ask, Attend, and Answer: A Strong Baseline For Visual Question Answering
• TAC-GAN - Text Conditioned Auxiliary Classifier Generative Adversarial Network

Cited by in corpus (3)

• Adversarial Text-to-Image Synthesis: A Review
• ALR-GAN: Adaptive Layout Refinement for Text-to-Image Synthesis
• Adma-GAN: Attribute-Driven Memory Augmented GANs for Text-to-Image Generation