Logo Generation Using Regional Features: A Faster R-CNN Approach to Generative Adversarial Networks

TL;DR

This paper presents LL-GAN, a logo generation model that leverages regional features from Faster R-CNN, achieving superior quality metrics on a heavy metal logo dataset compared to existing models.

Contribution

The paper introduces a novel logo generation framework combining Faster R-CNN regional features with GANs, outperforming state-of-the-art models on style-rich logo datasets.

Findings

LL-GAN achieves an Inception Score of 5.29

LL-GAN attains a FID of 223.94

Outperforms StyleGAN2 and Self-Attention GAN

Abstract

In this paper we introduce Local Logo Generative Adversarial Network (LL-GAN) that uses regional features extracted from Faster R-CNN for logo generation. We demonstrate the strength of this approach by training the framework on a small style-rich dataset of real heavy metal logos to generate new ones. LL-GAN achieves Inception Score of 5.29 and Frechet Inception Distance of 223.94, improving on state-of-the-art models StyleGAN2 and Self-Attention GAN.