A Novel Visual Representation on Text Using Diverse Conditional GAN for Visual Recognition

Automatic image visual recognition can make full use of largely available images with text descriptions on social media platforms to build large-scale image labeled datasets. In this paper, we propose a novel visual text representation, named DG-VRT (Diverse GAN-Visual Representation on Text), which extracts visual features from synthetic images generated by a diverse conditional Generative Adversarial Network (DCGAN) on the text, for visual recognition. The DCGAN incorporates the current state-of-the-art text-to-image GANs and generates multiple synthetic images with various prior noises conditioned on a text. Then we extract deep visual features from the generated synthetic images to explore the underlying visual concepts and provide a visual transformation on text in feature space. Finally, we combine image-level visual features, text-level features and visual features based on synthetic images together to recognize the images, and we also extend the proposed work to semantic segmentation. We conduct extensive experiments on two benchmark datasets and the experimental results demonstrate the efficacy of our proposed representation on text for visual recognition.

I. Introduction

Nowadays, images are being taken and shared to be commented on an unprecedented rate among social networks like Facebook, Twitter, and Flickr. Images in these social media platforms do not exist in isolation and most images on the web carry rich text information including informative and semantic signals like who takes the photo, and where and with whom. Therefore, it is desirable to explore social media context, especially text context information jointly with pixel information, to aid visual recognition tasks on images.