We study the joint learning of image-to-text and text-to-image generations, which are naturally bi-directional tasks. Typical existing works design two separate task-specific models for each task, which impose expensive design efforts. In this work, we propose a unified image-and-text generative framework based on a single multimodal model to jointly study the bi-directional tasks. We adopt Transformer as our unified architecture for its strong performance and task-agnostic design. Specifically, we formulate both tasks as sequence generation tasks, where we represent images and text as unified sequences of tokens, and the Transformer learns multimodal interactions to generate sequences. We further propose two-level granularity feature representations and sequence-level training to improve the Transformer-based unified framework. Experiments show that our approach significantly improves previous Transformer-based model X-LXMERT's FID from 37.0 to 29.9 (lower is better) for text-to-image generation, and improves CIDEr-D score from 100.9% to 122.6% for fine-tuned image-to-text generation on the MS-COCO dataset. Our code is available online.

本文研究了图片到文本和文本到图片生成的联合学习，使用了基于Transformer 的单个多模式模型来统一学习双向任务，并通过两级粒度特征表示和序列级训练进一步改进Transformer的统一框架，实验证明，该方法显著提高了先前基于Transformer的模型X-LXMERT的文本到图像生成的FID（37.0→29.9）和在MS-COCO数据集上对细调图像到文本生成的CIDEr-D得分（100.9%→122.6%）。

双向图像与文本生成统一多模态转换器