cycle text2face: cycle text-to-face gan via transformers

TL;DR

Cycle Text2Face introduces a novel encoder-decoder GAN model utilizing transformers for detailed text-to-face generation, achieving superior quality and speed compared to previous models on the CelebA dataset.

Contribution

The paper presents a new encoder-decoder architecture combining sentence transformers and GANs for text-to-face synthesis, with cycle consistency for improved results.

Findings

Achieved an FID score of 3.458 on CelebA.

Outperformed previous GAN-based text-to-face models.

Provided high-speed face image generation.

Abstract

Text-to-face is a subset of text-to-image that require more complex architecture due to their more detailed production. In this paper, we present an encoder-decoder model called Cycle Text2Face. Cycle Text2Face is a new initiative in the encoder part, it uses a sentence transformer and GAN to generate the image described by the text. The Cycle is completed by reproducing the text of the face in the decoder part of the model. Evaluating the model using the CelebA dataset, leads to better results than previous GAN-based models. In measuring the quality of the generate face, in addition to satisfying the human audience, we obtain an FID score of 3.458. This model, with high-speed processing, provides quality face images in the short time.