In this paper we explore to what extent machine learning can solve the creative task of the designer. For this, we build a dataset -- LLD -- of 600k+ logos crawled from the world wide web.

Dec 12, 2017 · Interpolation between 4 selected logos of distinct classes using DCGAN-LC with 100 AE clusters on LLD-icon, showcasing smooth transitions and interesting intermediate …

Code for reproducing experiments in "Logo Generation and Manipulation with Clustered Generative Adversarial Networks". The models are mainly meant to work with data in HDF5 …

The LLD-icon dataset is a subset of the LLD dataset, containing 486,377 logos in 32x32 pixel resolution.

Mar 7, 2024 · The LLD dataset consists of two parts, a low resolution (32 × \times 32 pixel) favicon subset (LLD-icon) and the higher-resolution (400 × \times 400 pixel) twitter subset …

In Figure 1 we present an example interface of a logo generator which could be used to facilitate the design pro-cess of a new logo. The samples used here where generated by our iWGAN …

Apr 12, 2022 · We used the Large Logo Dataset (LLD) from Logo Synthesis and Manipulation with Clustered Generative Adversarial Networks by Sage et al. 2017. The LLD-icon subset …

This paper analyzes a large number of logo images from the LLD-logo dataset, by recent deep learning-based techniques, to understand not only design trends of logo images and but also …

Jan 7, 2022 · LLD comes in two forms, LLD-icon and LLD-logo. As the existing benchmark adopted the icons set [22] , we follow the same route to design and implement our …

LLD-Icon color icon generation rendering based on ACWGAN. This paper focuses on exploring the application possibilities and optimization problems of Generative Adversarial Networks …